Skip to main content

Full text of "Heraclitus"

See other formats








Copyright © 1959 by Princeton University Press 

London : Oxford University Press 

All Rights Reserved 

L. C. Card No. 59-11086 

Printed in the United States of America by 
Princeton University Press 


my first teacher 

Elizabeth T>. <J)(Ceeker 

in gratitude 

Digitized by the Internet Archive 

in 2012 with funding from 

LYRASIS IVIembers and Sloan Foundation 


"There is a harmony in the bending back, as in the case 
of the bow and the lyre." Thus HeracHtus observes in Frag- 
ment 117; and the complementing virtues of marksmanship 
and lyrical sensitivity must both be courted in any attempt 
to rediscover and freshly interpret the extraordinary subtle- 
ties and depths of his ancient wisdom. 

The present book was begun during a special leave of 
absence granted me by the University of California at River- 
side; it was brought to virtual completion while I was the 
visiting William Allan Neilson research professor at Smith 
College. I am grateful to both institutions for the oppor- 
tunities thus provided. 

Five professorial friends have generously put their scholar- 
ship at my disposal as needed: Professor Harry Carroll of 
Pomona College; Mr. W. Dyfrig Evans, while he was a 
visiting member of the U.C.R. Classics Department; Profes- 
sor Alfred Young Fisher of Smith College ; Professor Philip 
Merlan of Scripps College; and Professor Wolfgang Your- 
grau, at present the Cowling Visiting Professor of Philosophy 
at Carleton College. The last named (last only by alpha- 
betical accident) has brought his special knowledge of the 
philosophy of science and his familiarity with Greek lan- 
guage and literature jointly to bear upon the problems of 
ancient science discussed in Chapter III. I am indebted no 
less, although in different respects, to Miss Harriet Anderson 
of the Princeton University Press, for her constant exercise 
of sound judgment, good taste, and technical proficiency 
during the months of our collaboration; to Dr. Gertrud 
Neuwirth of the University of Frankfurt for varied assist- 
ance; to Miss Dorothea Berry, reference librarian at U.C.R. ; 
and to my wife, who has labored long with the proofs at 
their successive stages. 

Finally, my gratitude is real although less specifically 
focussed toward those students in my ancient philosophy 

[ vii ] 


courses, recently at the University of California at Riverside 
and earlier at Dartmouth College, who have shared in the 
excitement of rediscovering Heraclitus with me, and whose 
responsiveness has helped to confirm my abiding conviction 
of the philosopher's greatness. 

P. W. 

University of California 
at Riverside 
August ip^p 

[ viii ] 




















The sixth century B.C. was a time of philosophical ferment 
in many places. If we do not draw the boundaries of the cen- 
tury too exactly, we may include within it, with a fairly high 
degree of probability, such widely separated and often diverse 
philosophers as Lao-tze in China, Zoroaster (more properly 
Zarathustra) in Iran, the anonymous authors of several of 
the Upanishads in India, the Deutero-Isaiah in Exilic Israel, 
and a number of different and sometimes sharply opposed 
individualists in various parts of the Greek-speaking world. 
It is a mystery, which no one has succeeded in explaining, 
why there should have been such a nearly simultaneous 
concentration of philosophical activities in countries that 
appear to have had no direct connection with each other. At 
all events, our present concern is with one particular aspect 
of this general manifestation; and the reader may observe 
or ignore, at his discretion, the somewhat analogous develop- 
ments occurring at the same time in different parts of the 
then civilized world. 

So far as Greece is concerned, there are virtually no evi- 
dences of anything that could properly be called philosophy 
existing in earlier times. There are momentary flashes of 
philosophical insight in Hesiod's Works and Days, in the 
so-called Homeric Hymns (whatever their dates may have 
been), and in the fragmentary sayings attributed to Orpheus 
and his followers; but in none of them is there any intellectual 
coherence _Qr any interest in finding a method for distinguish- 
ing truth from error. The ancient saying cited by Plato, that 
''God holds the beginning and the end, as well as the middle, 
of all existing things,"^ has been attributed to the legendary 
Orpheus, and it contains an arresting thought — especially if, 
as was usual with the Greeks, the ideas of beginning, end, 
and middle carried not only temporal but also moral mean- 
ings, connoting respectively (i) principle, (2) goal or ful- 
fillment, and (3) balance or proportion. But the quotation 

[ 3 ] 


stands virtually alone ; most of the other early Orphic sayings 
are more burdened with mythological assumptions, and none 
of them reveals so clear an intellectual focus. In the gnomic 
poets of the seventh century, too, there are a few scattered 
utterances that show flashes of philosophic insight, but here 
again the occasional sense of significance is sporadic, un- 
methodical, and unpromising. 


The first independent and sustained attempt to work out a 
philosophic view of the world is found in Ionia, to the east 
of Greece proper, injhe sixth century. Here, in the seaport 
town of Miletus, Thales and his two successors, Anaximander 
and Anaxinienes, began to ask questions in a new way. De- 
scribed by ancient writers as *'physiologues,"^ because they 
were seeking a reasoned understanding {logos) of nature 
(physis), they formulated — apparently for the first time in 
the Western world — the two great scientific-philosophical 
questions of What and How. The two things they chiefly 
wanted to know were : 'What is the primary stuff of which 
the world is constituted?" and "How do the changes take 
place that bring about its manifold appearances?" Previously 
the nearest that the Greek mind had come to launching such 
inquiries was to ask not *'What?" but "Who?" and not 
"How?" but "With what purpose and intention?" The his- 
torical importance of Thales, as founder of the Milesian 
school, does not rest on his somewhat primitive and naive 
theory that all things are transformations of water, nor on 
the legends of his wizardry and absent-mindedness, but on 
the fact that he began the work of seeking for explanations 
of the natural world within the natural world itself. An un- 
derstanding and formulation of the two questions — "What 
is basically real?" and "How does change come about?" — is 
far more important, both for clarity of mind and in terms of 
subsequent influence, than any possible answers that can 
ever, then or now, be given to them. 

[ 4 ] 


Thales' greatest follower, Anaximander, made an attempt 
to answer the first of these questions by his conception of a 
boundless reservoir of potential qualities,^ out of which 
warmth or coldness, health or sickness, light or dark, and so 
on, would emerge into actual existence at certain times and 
places and subsequently would be reabsorbed into that bound- 
less cosmic reservoir (to speak, as one must, metaphorically) 
in which all things exist merely as potentials. Anaximander 
answered the second question by his conception — partly bio- 
logical, partly ethical, partly religious — of what can most 
nearly be described as existential penance. 'Things make 
reparation," he declared, ''and therein do justice to one an- 
other according to the order of time."* This is his one pre- 
served statement of how a change from one quality to another 
— say from warmth to coolness in the atmosphere — comes 
about. The meaning can best be understood by looking at it 
in two perspectives successively. In biological perspective we 
can — on the analogy of an organism that grows, reproduces, 
and dies — regard a quality, such as summer heat, as coming 
into being, achieving full growth, and then, after a suitable 
time, making way for the opposite quality — in this case winter 
cold — which is to be conceived as going in its turn through 
the same life-cycle. In ethico-religious perspective the situa- 
tion can be conceived through the typically Greek idea of 
hybris, vjhioh can be roughly translated flagrant self-asser- 
tion. Now what Anaximander's metaphysical imagination 
has done is to envisage the process of flagrant self-assertion 
together with its self-terminating outcome as applying not 
only to human life but to all existing entities whatsoever. The 
light of day, when it has asserted itself by existing for enough 
hours, must at length yield to the conflicting claims of the 
darkness of night, which has been lurking in a merely poten- 
tial state, awaiting, as it were, its chance of bursting into 

Anaximander's doctrine, so far as it deals with the prob- 
lem of change, is a forerunner of Heraclitus' doctrine in two 

[ 5 ] 


respects. In the first place it conceives of change in purely 
qualitative terms. That is to say, change is not, as it is for 
modern physical science, primarily spatial motion, or even 
measurable by some device of spatial motion or spatial dis- 
tancing. It is essentially what it appears to be — the disappear- 
ance of one perceptual quality (say felt warmth or visible 
brightness or audible noise) while another and contrary 
quality (respectively, coolness, darkness, or silence) takes its 
place. Change is an ontological passage from contrary to con- 
trary — ^from one perceptible state of being to its opposite. In 
the second place Anaximander conceives of the relation be- 
tween contraries as in some sense a periodic interchange, 
exemplifying a cyclic and somewhat rhythmic principle — 
"according to the order of time." 

Anaximenes, too, the last of the three Milesian philoso- 
phers, although generally regarded as inferior to his im- 
mediate predecessor in philosophical stature, nevertheless 
offers two teachings that make him a significant forerunner 
of Heraclitus. First, he takes the primary physical reality to 
be air; for, as he argues in one of his few surviving frag- 
ments, "As our souls, being air, hold us together, so do 
breath and air encompass the entire universe."^ His second 
thesis — of great historical importance in that it opened the 
way to the later employment of quantitative concepts in 
physical science — is the interpretation of change in terms of 
serial order. Every occurrence in nature, he holds, is a result 
and outward show of the rarefaction or condensation of air. 
Rarefaction eventually produces fire; condensation at suc- 
cessive stages produces cloud, water, mud, earth, and rock. 
Here evidently is something roughly analogous to Heraclitus' 
"way up and way down," and the question suggests itself 
whether Heraclitus may have been influenced by Anaximenes, 
or whether both philosophers may have drawn upon a com- 
mon source ; but unfortunately there is no evidence on which 
an answer can rest. 

One further Ionian philosopher, not from Miletus and 

[ 6 ] 


quite distinct from the trio just mentioned, invites considera- 
tion because of his possible bearing upon the thought of 
HeracHtus. Xenophanes was born in the Ionian town of 
Colophon, probably about a generation before HeracHtus, and 
he appears to have spent much of his time as a traveling min- 
strel, taking as the themes of his songs philosophical and 
religious ideas. Although his utterances are of a more open 
kind, less gnomic and paradoxical, than those of HeracHtus, 
there is a certain eloquent bitterness to be found in both 
philosophers, particularly as directed against prevailing 
stupidities of belief. In Xenophanes' case the doctrine that is 
especially under attack is the popular mythology of the 
Olympian gods. The gods, taken plurally, he denounces as 
fictions in anthropomorphic guise. If horses could draw, he 
remarks, they "would portray their gods in the shape of 
horses, and oxen in the shape of oxen." We are victims, he 
could have said, using Bacon's later figure, of idols of the 
tribe. Is there any escape from such idols? Yes, Xenophanes 
thinks, there is. We must practice the art of ridding our 
highest conceptions of all accidental, trivial, and self-mirror- 
ing qualities. As a result he arrives at the first clear statement 
of monotheism, so far as is known, in the West. ''There is one 
God," he declares, *'the greatest among gods and men, not at 
all like mortals in form or in thought." To be sure, since we 
regard this supreme One as divine, we must conceive Him 
as more excellent, not less so, than mortals, and this means 
that He must be somehow capable of thought and perception, 
for an entirely unconscious entity would not be worthy of 
reverence. Such thought and perception must not be con- 
ceived as dependent on organs, however; ''it is in his entirety 
that he sees, in his entirety that he thinks, and in his entirety 
that he hears." Nor does he have to exert himself in order 
to bring things about; "he accomplishes everything by the 
sheer thought of his mind."® He is, in short, the transcendent 
unifier and the principle of unity that resides amidst all change 
and multiplicity. Xenophanes has thus taken an important 

[ 7 ] 


and apparently an original step in setting up the concept of a 
God divested of human attributes. His critical approach to 
philosophy through the religious problem complements the 
Milesian critical approach through the problem of nature. 
Metaphysics, or what we may call more precisely cosmology, 
involves both approaches ; they provide respectively the polar 
concepts of the One and the Many. 


Two other philosophies, the Pythagorean and the Eleatic, 
need to be mentioned as partly contemporaneous with Hera- 
clitus, and as just possibly having had some oblique influence 
upon him, although this is doubtful. Granted that both 
philosophies reached their peak of influence after the probable 
date of Heraclitus' death, nevertheless in their earlier phases 
they may perhaps have been known to him in indirect ways. 
I am not arguing that this was so, but it is just as well to 
reckon with the possibility. 

There is certainly no evidence or likelihood that Heraclitus 
had ever been influenced by Pythagoras directly. The two 
derogatory remarks, of doubtful authenticity, which he was 
later quoted as having made about the older philosopher may 
be found as Frs. 128 and 136 in Appendix C. If he actually 
did make them, it must be admitted that he showed neither 
understanding nor sympathy toward Pythagoras and his 
teaching. Most of Pythagoras' teaching, moreover, was done 
in Italy, in the little city of Crotona where he had established 
a brotherhood for the further pursuit of philosophy; and 
although he was probably older than Heraclitus by a genera- 
tion, it is improbable that his views, which were carefully 
guarded, would have been carried to Ionia within a few years 
or decades. Finally, the two men, to judge by the available 
evidence, were sharply different in intellectual temperament. 
Pythagoras' doctrine, compounded of mysticism, mathe- 
matics, cosmology, and music, is so alien to our current 
preconceptions that its original intelligibility is difficult to 

[ 8 ] 


recapture; and while Heraclitus would not have opposed it 
for the same reasons as we, he was too fiercely individualistic 
to accept any such teaching from another, even if he had come 
in contact with it. 

On the other hand, there is an ancient tradition that Hera- 
clitus had been a pupil of the Pythagorean philosopher Hip- 
pasus of Metapontum. The tradition is too shaky to stand as 
evidence, however; for some scholars regard Hippasus as 
having postdated Heraclitus, and moreover lamblichus in his 
Lije of Pythagoras discredits his Pythagoreanism. The two 
firm facts about Hippasus are that he had been at one time 
a member of the Pythagorean brotherhood, and that like 
Heraclitus he believed that the universe is in a state of in- 
cessant change and that it consists of fire as its primal 
element/ But whether he held the belief ahead of HeracHtus 
or drew it from him, is impossible to determine; and in any 
case so simple and general a congruity might, of course, have 
been coincidental. 

The problem of Heraclitus' relation to Parmenides of Elea 
again involves an uncertain question of comparative dates. 
A majority of scholars have supposed that Parmenides wrote 
his long poem on truth and appearance after Heraclitus had 
published his treatise on nature. Two main reasons for the 
conventional dating may be noticed. First, there is the evi- 
dence of Plato's dialogue, the Parmenides^ in which Socrates 
as a young man is represented as conversing with the aged 
Parmenides. Since Socrates was born in 469 B.C., the con- 
versation (assuming that Plato did not invent it) would 
probably have taken place not earlier than about 450. In that 
case, whatever Parmenides' exact age may have been at the 
time, it would seem probable that the writing of his philosophy 
would have taken place after, rather than before, the turn of 
the century, which was the time at which Heraclitus is said 
to have flourished. Secondly, there is the passage supposed 
by a number of scholars to refer to Heraclitus, in which 
Parmenides warns against "undisciplined crowds who hold 

[ 9 ] 


that to be and not to be are the same and yet not the same, 
and that the way of things everywhere is TraXti^rpoTro?."^ Al- 
though, as Zeller has pointed out, Heraditus does not in any 
extant Fragment say that being and not-being are the same 
and not the same (and, incidentally, he can hardly be well 
described as an undisciplined crowd), yet the word Trakiv- 
TpoTTO'^ (^'bending back," ''tension between opposites") is so 
distinctive as to suggest the possibility of a deliberate refer- 
ence to Fragment 117, where Heraclitus uses the same word. 
Nevertheless, despite these rather strong evidences, it is 
not certain that Parmenides' writing postdates that of Hera- 
clitus. A scholar of independent mind, Karl Reinhardt, has 
argued with much learning and ingenuity that Heraclitus 
may have flourished as much as two decades after Par- 
menides, and that his theory of an ever-changing universe 
arose in reaction against Parmenides' unworkable theory that 
change is unreal.® Parmenides had argued that an intelligible 
view of the world is only possible to one who regards it as 
undifferentiated and static; Heraclitus (if Reinhardt's re- 
dating is accepted) could be interpreted as retorting to Par- 
menides' oversimple theory with the countertheory that in- 
telligibility is to be found only in what is multiple and chang- 
ing — only in strife itself. To be sure, Reinhardt's dating has 
not won wide acceptance; nevertheless the possibility of 
Heraclitus' having reacted against Parmenides in the manner 
described should doubtless be kept open. 


Heraclitus himself was a native of Ephesus, an Ionian city 
some twenty-five miles north of Miletus and inland from the 
sea, and he is said by Diogenes Laertius to have flourished 
there in the sixty-ninth Olympiad, which would be roughly 
equivalent to 504-500 B.C. His family was an ancient and 
noble one in the district, and Heraclitus inherited from them 
some kind of office, partly religious, partly political, the exact 
nature of which is not clear, but it involved among other 

[ 10 ] 


things supervision of sacrifices. Doubtless such an office was 
not congenial to a man of his impatient temperament, and he 
resigned it in favor of a younger brother. The banishment 
of his friend Hermodorus by a democratic government in- 
creased a natural antagonism to the masses and confirmed him 
in his philosophical withdrawal. So much is virtually all that 
can be known about Heraclitus with reasonable probability. 
Diogenes Laertius' short essay on him in Lives and Opinions 
of Eminent Philosophers^^ is a rather scatterbrained affair, 
and there is no reason to take seriously his fantastic account 
of the philosopher's death by self-burial in a cow stall in a 
vain effort to cure an attack of dropsy. Such improbable tales 
were not uncommon about ancient "wise men," and Diogenes 
provides more than his share of them; quite possibly their 
origin was aetiological in that they grew out of popular mis- 
understandings of something that the philosopher had taught. 
In the case of Heraclitus we cannot even know whether it is 
true that he died of dropsy ; the story could easily have been a 
figment suggested by his remark, "It is death for souls to 
become water." 

In temperament and character Heraclitus was said to have 
been gloomy, supercilious, and perverse. Diogenes calls him 
a hater of mankind, and says that this characteristic led him 
to live in the mountains, making his diet on grass and roots, a 
regimen which brought on his final illness. Such an account, 
however, is of the sort that could easily have been invented 
out of a general view of the philosopher's character. At any 
rate, Heraclitus was certainly no lover of the masses, and his 
declaration, "To me one man is worth ten thousand if he is 
first-rate" (Fr. 84), makes it evident that he was not one to 
suffer fools gladly. He would have understood and approved 
of Nietzsche's definition of the truly aristocratic man as one 
whose thoughts, words, and deeds are inwardly motivated 
by a "feeling of distance. "^^ However, to call him a pessimist 
and compare him to Schopenhauer, as more than one inter- 
preter of his writings has done, is to treat him in a mislead- 

[ II ] 


ingly one-sided manner. Pessimism, where it is a philosophy 
and not just a mood, affirms the doctrine that there is more 
evil in the world than good, or that the evil is somehow more 
fundamental or more real. Heraclitus does not commit him- 
self to so partisan a statement. His doctrine is rather that 
good and evil are two sides of the same reality, as are up 
and down, beauty and ugliness, life and death. The wise man 
attempts to set his mood by looking unflinchingly at both sides 
of the picture, not at either the bright or the dark alone. 

So far as is known, Heraclitus was the author of a single 
book. Diogenes Laertius describes its subject-matter as "on 
nature," adding that it was divided into three sections — on 
the universe, on statecraft, and on theology. According to that 
same sketchy biographer Heraclitus dedicated his book in the 
temple of Artemis and deposited a scroll of it there — a fairly 
usual practice in ancient Greece.^^ The virtually unanimous 
opinion of ancient writers is that the book was hard to un- 
derstand, and its author was frequently described by such 
epithets as the Dark, the Obscure, and the Riddling. ^^ Diog- 
enes Laertius suggests that the obscurity may have been 
deliberate, in order that none might read the book who had 
not honored it with a suitable degree of intellectual effort. 
Since the Fragments that survive from it consist mostly of 
single sentences, we have little or no direct evidence as to how 
the various ideas in it were assembled, but their pithiness 
and profundity are still unmistakable, even though the original 
contexts have been lost. 


Since Heraclitus is one of the subtlest, most impatient, and 
most paradoxical of philosophers, any attempt to reduce his 
doctrine to a few plain propositions could only result in dis- 
tortion and caricature. Interpretations of his meaning have 
to be somewhat tentative, left open to the qualifications and 
reformulations that a thoughtful reader may wish to make as 
he reflects further on some of the dark sayings. There is 

[ 12 ] 


always the danger, with Heraclitus as with any other ancient 
philosopher, of interpreting the ideas and the words — or 
someone's chosen translation of the words — on the basis of 
contemporary presuppositions and distinctions, such as may 
have been absent, or nearly so, from the thought of an earlier 
age. We become subject, more than we are aware, to idols 
of the theater/* In particular there are three modes of dis- 
tinguishing, which seem quite natural to us today, but which 
are relied on to a far lesser degree in the thought and expres- 
sion of Heraclitus : our grammatical distinction among parts 
of speech, our logical distinction between the concrete and the 
abstract, and our epistemological distinction between subject 
and object. 

The distinction among parts of speech is less pronounced in 
the Greek language than in the Latin and its Western suc- 
cessors. Accordingly it is often impossible for a translator to 
find in a modern language the precise equivalent of some word 
or idiom in the Greek. The difficulty, indeed, is more than 
grammatical, it is ontological; for it concerns the kind of 
being which the different types of words are designed to 
indicate. Of particular interest are the three word-types of 
noun, adjective, and verb, together with the three modes of 
being for which they respectively stand — things, qualities, and 
events. The correlation is not absolute, to be sure ; for we have 
to remind ourselves occasionally that an abstract noun such 
as "justice" does not indicate a thing, and that the copula 
*'is" lacks the usual semantic properties of other verbs. But 
in general our contemporary Western languages keep a 
fairly steadfast distinction among the three types — nouns 
standing for things, adjectives standing for qualities, and 
verbs standing for actions and events. 

Now in the thought of Heraclitus, abetted by the com- 
parative fluidity of the Greek language, the linguistic distinc- 
tion and correspondingly the ontological distinction are some- 
what less firm. Consider Fr. 22, for instance: ''Cool things 
become warm, the warm grows cool; the moist dries, the 

[ 13 ] 


parched becomes moist." Is Heraclitus speaking here about 
things or about qualities? The subject of the first clause is 
neuter plural preceded by the definite article, while in each of 
the three remaining clauses the subject term is a neuter singu- 
lar without the article. Yet it is evident that the four clauses are 
parallel, and that the four subject terms are offered as repre- 
senting parallel situations. The answer seems to be that 
scarcely any distinction was recognized between cool and 
warm things and the resident qualities of coolness and 
warmth. It was not until a century and a half later that 
Aristotle delineated the difference explicitly and showed what 
the intellectual penalty for ignoring it would be — a point to be 
developed in Chapter II. And in Heraclitus' thought not only 
the ideas of thing and quality but also those of event and 
quality tend to coalesce and become confused. The latter con- 
fusion tends to be encouraged by the readiness of the Greek 
language to employ the infinitive of a verb preceded by the 
neuter definite article. Where such a construction appears in 
Fr. lo, for instance, I have translated it "to be temperate," 
although a more literal translation would be 'Hhe to be 
temperate," and doubtless the sense of action (or potential 
action), quality, and thinghood were all present in it. 

The coalescence between concrete and abstract is especially 
evident in Heraclitus' central image-idea of fire. Regarding 
Frs. 28, 29, 30, and 32, the question has been raised: Is he 
speaking about actual physical fire, which burns and flares, or 
is he employing a picturesque symbol to denote incessant 
change? No one-sided answer can be maintained without 
doing violence to the doctrine; the true answer has to be — 
both ! Goethe, who evinced a lively interest in Heraclitus after 
Schleiermacher had presented him with a first collection of 
some of the Fragments, defines a genuine symbol as a par- 
ticular instance which is coalescent with a universal and which 
thereby plays a unique role by revealing, in a way that no 
other particular could quite do, the nature of that more gen- 
eral something.^^ Fire, in Heraclitus' doctrine, is a symbol 

[ 14 ] 


in something very like the Goethean sense. It is the yellow, \ 
flaming, heat-giving actuality while at the same time it stands I 
for the Heraclitean principle of universal unrelenting change. ' 

The third type of coalescence that is more natural to earlier 
methods of thinking than to our own is found in the coales- 
cence of subject and object, or knower and known, or thinker 
and thing thought. A comparison of Frs. 119 and 120 is in- 
structive in this regard. Each of these Fragments begins with 
the same three words, stating that wisdom (more literally, the 
quality of being wise) is one, but they amplify the idea in 
different directions. Fr. 119 identifies this unitary wisdom 
with the divine power that is active in all things, and 
thus has a somewhat objective reference; Fr. 120 identi- 
fies the same unitary wisdom with the power of knowing 
that cosmic intelligence, and thereby has a somewhat sub- 
jective reference. Such is the distinction that tends to suggest 
itself to the modern reader, one of whose most complacent 
assumptions is that he knows how to draw the line between 
what is objective and what is subjective, and that except in 
certain rare and abnormal cases there is no difficulty in doing 
so. To an ancient thinker, on the other hand, whose mind 
would not have been conditioned (as ours has largely been) 
by the postulates of Cartesian dualism, the division between 
objective and subjective wore no such appearance of clarity 
and finality. The idea of what might belong to the one and 
what might belong to the other would vary according to 
mood and circumstance, and no precise question about it was 
ever raised. Heraclitus shows at times (as in Frs. 11, 13, 
15, and 16) a strong, but not a clear, feeling for the ques- 
tion; Parmenides acknowledges its importance by building 
his poem upon a contrast between the way of truth and the 
way of opinion ; but it was not until the age of the Sophists 
that Greek Philosophy had acquired the vocabulary and the 
dialectical skill to handle the question firmly. The Sophists' 
own answer to it erred on the side of excessive subjectivity, 

[ 15 ] 


and Plato in attempting to correct their error produced a 
certain amount of metaphysical muddle. Aristotle's ingenious 
and promising attempt to solve the riddle^^ is disappointingly 
incomplete, and in general it may be said that Greek philos- 
ophy never succeeded in grappling with the problem ade- 
quately. Perhaps it is just as well. A seeker after truth will 
fare best if he tries, whatever the intellectual inconveniences, 
to keep his categories and his ontological premises somewhat 
flexible. Such flexibility is what gives to much early Greek 
philosophy, and particularly to that of Heraclitus, its charac- 
teristic resilience and vitality. 

These three kinds of semantic coalescence are not peculiar to 
the thought of the Greeks. With variations in detail they are 
to be found in much of the philosophical thought of ancient 
India, Iran, and China too. Because of them many of the 
remarks in the Upanishads, the Zend-Avesta, and the Tao 
Teh Ching appear dark and riddling, or on the other hand 
naive and superficial. In their clarity of thought, their sharp- 
ness of imagery, and their economy of ontological assump- 
tions, the sayings of Heraclitus are more akin to the Tao 
Teh Ching than to the religio-philosophical documents of 
India and Iran, although they are doubtless more like these 
in their prophetic tone. In any case the lucidity that charac- 
terizes most of Heraclitus' sayings (despite the ancient cliche 
to the contrary) must be understood on its own terms. It is 
sometimes subtly different from what passes for lucidity at 
the present day, and such difference is at bottom a difference 
in certain basic thought-forms, of which I have indicated three 
of the most prominent. A first step in trying to understand 
any writer, and particularly an ancient writer, is to exercise 
our "negative capability,"^^ bracketing off our habitual ways 
of joining and distinguishing ideas wherever such ways differ 
from the writer's own. This step is especially demanded by the 
flowing and often paradoxical thought of Heraclitus, and the 
cost of ignoring it would be a misleading oversimplification. 

[ i6 ] 



The problem of resurrecting the thought of an ancient 
writer whose remains consist only of quoted fragments will 
have two main phases. There is the question of the authen- 
ticity of particular fragments, and there is the question of 
how to interpret the fragments that are accepted. Once the 
authenticity of a fragment has been established, or at least 
accepted upon reasonable grounds, the question of interpre- 
tation may arise in any or all of several forms. Sometimes 
there is the problem of exactly what certain words or phrases 
meant in the late sixth century B.C. : such doubts, when 
troublesome, are discussed in Appendix B, the textual notes 
on the Fragments. Sometimes a modern reader's understand- 
ing is hampered by the philosophical, semantic, and gram- 
matical ambiguities discussed above: a difficulty for which 
there is no secure remedy, but which demands constant alert- 
ness of judgment and responsiveness to context. Then again, 
false scents are sometimes introduced by the later ancient 
writers who quote the fragments putting their own interpreta- 
tions upon them. Examples of this difficulty and suggested 
ways of meeting it are offered and discussed in Appendix B. 

Questions both of authenticity and of interpretation have 
been explored with much industry, scholarship, and ingenuity 
from the time of Schleiermacher's first compilation of the 
extant sayings of Heraclitus, in 1817, down to the present 
day. By general agreement among contemporary scholars the 
edition of Hermann Diels as revised by Walther Kranz in 
1934 is now taken as standard; and the Fragments of Hera- 
clitus that are presented in the following chapters are trans- 
lated from the Diels-Kranz text, except where otherwise 
noted in Appendix B. Five Fragments (20, 27, 43, 62, 74) 
omitted from the Diels-Kranz canon are included here in 
parentheses, for reasons that Appendix B explains and seeks 
to justify. The chapters themselves, following the presenta- 

[ 17 ] 


tion of the Fragments, are meant to be suggestive and per- 
spectival, not exhaustive or definitive. For the utterances of 
Heraditus, Hke those of the darkly luminous Lord at Delphi 
(Fr. i8), do not simply speak nor yet simply conceal; they 
''give signs." 

[ i8 ] 


^i. Although thisLo^osis eternally valid, yet men are unable 
to understand it — not only before hearing it, but even 
after they have heard it for the first time. That is to say, 
although all things come to pass in accordance with this 
Logos, men seem to be quite without any experience of 
it — at least if they are judged in the light of such words 
and deeds as I am here setting forth. My own method is to 
distinguish each thing according to its nature, and to 
specify how it behaves; other men, on the contrary, are as 
forgetful and heedless in their waking moments of what is 
going on around and within them as they are during sleep. 

^2. We should let ourselves be guided by what is common to all. 
Yet, although the Logos is common to all, most men live 
as if each of them had a private intelligence of his own. 
J. Men who love wisdom should acquaint themselves with a 
great many particulars. 

4. Seekers after gold dig up much earth and find little. 

5. Let us not make arbitrary conjectures about the greatest 

6. Much learning does not teach understanding. 

7. Of those whose discourses I have heard, there is not one 
who attains to the realization that wisdom stands apart 
from all else. 

* 8. 1_have searched myself. 
p. It pertains to all men to know themselves and to be tem- 

10. To be temperate is the greatest virtue. Wisdom consists in 
speaking and acting the truth, giving heed to the nature of 

11. The things of which there can be sight, hearing, and learn- 
ing — these are what I especially prise. 

12. Eyes are more accurate witnesses than ears. 

[ 19 ] 


/J. Eyes and ears are had witnesses to men having barbarian 

14. One should not act or speak as if he were asleep. 

15. The waking have one world in common; sleepers have 
each a private world of his own. 

16. Whatever we see when awake is death; when asleep, 

ly. Nature loves to hide. 

18. The lord whose oracle is at Delphi neither speaks nor con- 
ceals, but gives signs. 

ip. Unless you expect the unexpected you will never find 
[truth], for it is hard to discover and hard to attain. 

Every serious approach to philosophy must begin, whether 
expHcitly or not, with some consideration of method. In cer- 
tain of the most famihar cases the matter is indicated by 
indirection. Plato shows what is most essential to his method 
by the dialogue form in which his philosophy is clothed. Most 
of Aristotle's treatises reveal his concern for order by be- 
ginning with a universal : ''All men by nature desire to know" 
(Metaphysics) , ''In any inquiry that has to do with principles 
or causes or elements" (Physics), "Every art and every 
science . . . aims at some good" (Ethics). Among philos- 
ophers since the Renaissance there has been somewhat more 
of a disposition to formulate methodological postulates ex- 
plicitly : in particular, Bacon and Locke on the one hand, Des- 
cartes, Spinoza, and Kant on the other, have given deliberate 
attention to the problem ; and in a curiously oblique way even 
Berkeley's New Essay towards a Theory of Vision offers, as 
a corollary of its optical discoveries, a clue to the method that 
was to produce the central paradox most associated with the 
philosopher's name. Heraclitus, too, evidently opened his 
treatise with an explicit avowal of the method he intended to 
pursue : for according to the testimony of Sextus Empiricus, 
Frs. I and 2 originally stood, the one at the beginning of the 
work, the other a short distance further on.^ 

[ 20 ] 


A vexed question has arisen concerning the interpretation 
to be given to the word Xoyo^ as it appears in the first two 
Fragments. Granted that the literal meaning is something like 
''word," is the reference primarily to a transcendental and 
universal Other who speaks as it were, or primarily to the 
utterance that Heraclitus himself is making? Although the 
balance of evidence seems to support the former interpreta- 
tion, no less an authority than John Burnet has maintained 
the latter.^ 

The argument for a subjective interpretation of X0709 in 
Frs. I and 2 may be put as follows. From what is known 
about the practice of heading a treatise in early Greece, it is 
probable that in lieu of a title Heraclitus would have employed 
a formula such as : "Heraclitus of Ephesus, son of Blosson, 
speaks as follows."^ If he did so, and if we are to interpret 
Sextus as meaning that Fr. i occurred immediately after this 
conventional opening, then it would seem likely that the word 
X6yo<; in Fr. i was intended to pick up the idea in the verb 
"speaks" (Xeyet) which preceded it; so that on this ground 
Burnet would appear to be right in supposing that the word 
Xoyos in the Fragment refers primarily to the discourse of 
Heraclitus himself. 

There is a grave objection, however, to accepting so per- 
sonal an interpretation of the word. Sextus Empiricus, in 
quoting the passage, appears to interpret the word as bearing 
a universal and cosmic significance, for two sentences later he 
writes: "Heraclitus asserts that the common and divine 
Logos, by participation in which we become rational, is the 
criterion of truth."* Moreover, in Fr. 118, Heraclitus 
pointedly distinguishes between the Logos and his own per- 
sonal speech, when he says: "Listening not to me but to the 
Logos. . . ." Although Burnet's rendering, "It is wise to 
hearken, not to me, but to my word ..." can be defended 
on grounds that are indicated in the Note to Fr. 118, it 
nevertheless makes awkward sense, and seems to be a result 
of trying to remain consistent with the interpretation already 

[ 21 ] 


given to \6yo<; in Fr. i. But instead of starting out by de- 
ciding how to interpret Fr. i and then being obHged to inter- 
pret Fr. ii8 in so forced a manner, it would be a better 
procedure to keep both Fragments in mind when trying to 
discover what either of them means. As a matter of fact, 
Burnet's argument supporting his interpretation of Fr. i is 
not airtight, for neither of the premises on which it rests is 
quite certain beyond reasonable doubt. On the one hand, it 
is not certain how exactly Sextus' phrase, "at the beginning 
of his treatise," is to be understood. Nor, on the other hand, 
do we know for sure that the treatise was headed by the usual 
formula, ''Heraclitus speaks thus." Moreover, even if such a 
heading was actually employed, we cannot know how much 
of a prophetic connotation the word "speaks" may have 
carried. To a degree that our secular and colloquial habits of 
thought can scarcely comprehend, the ancient speaker was apt 
to regard his speaking not as a personal activity but as a 
voicing forth of a something greater — of a Logos whose 
character cannot be given, except haltingly and fragmentarily, 
in the human utterance. On this basis, then, if Burnet's two 
premises are accepted, it might still be possible to interpret 
the opening formula as signifying, "Heraclitus voices forth 
the Logos as follows" ; the reference of "this Logos" in Fr. i 
would thus be to the cosmic principle and not to the personal 

A somewhat analogous interpretation is pertinent to Fr. 2. 
"Although the Logos is common to all, most men live as if 
each of them had a private intelligence of his own." Burnet's 
version, "Although my word is common . . . ," besides lacking 
any justification in the Greek text, substitutes a confused and 
queer meaning for the natural one. What can be meant by 
saying that "my word" is common to all? Whatever meaning 
it might possess would seem to be at odds with Heraclitus' 
fiercely aristocratic temper. Once again, the clue for a reason- 
able interpretation is found in Sextus Empiricus. For after 
quoting Fr. i Sextus writes, "Having in these words argued 

[ " ] 


that we do and think everything through participation in the 
divine Logos. . . ." Then, after quoting Fr. 2, he goes on to 
remark: "And this is nothing else than an explanation of the 
way in which the entirety of things is arranged. Therefore, 
insofar as we share in the memory of this we say what is 
true, but when we depend on private experience we say what 
is false."^ Editors differ as to whether ''this," in the memory 
of which we may share, refers to ''the entirety of things" or 
to the word "Logos" in the sentence preceding. But in either 
case it seems clear enough that Sextus understands the word 
"Logos" as having a cosmic and universal sort of reference. 

An interesting light, although no direct evidence, is thrown 
on the problem by a fragment from a near contemporary of 
Heraclitus, one Epicharmus of Syracuse, who flourished 
about 480 to 470 B.C. He was a writer of comedies who occa- 
sionally ventured into philosophical speculation, and the fol- 
lowing passage from his philosophical writings is quoted by 
Clement of Alexandria. "The Logos," Epicharmus writes, 
"steers men and ever preserves them in the right way. Man 
has the power of calculation, but there is also the divine 
Logos. Human reasoning is sprung from the divine Logos, 
which furnishes to each man the passageway of both life and 

At any rate, as I have remarked in the Introduction, the 
solution to such questions is not to be found in terms of a 
sharp either-or. The strongest likelihood is that Heraclitus 
regarded the Logos as the Truth in its objective and trans- 
human character, and yet also regarded himself as being 
especially qualified and privileged to reveal the nature of that 
truth. To be sure, the word X0709 also means "word," and 
thus the connotation "what is spoken" attaches to it. But the 
connoted idea of speaking is largely metaphorical and tran- 
scendental; much as when nowadays, with somewhat less 
conviction, we employ the phrase, "the voice of truth." In 
Heraclitus' day the sage {cro<^6'^) or lover of wisdom 
(^L\6cro(j)o<;) was still the man who found himself called, by 

[ 23 ] 


the voice of a presence greater than he, to speak forth the 
truth as it might be revealed to him. 

What does HeracHtus mean by describing the truth as 
''common"? The meaning is put accurately by Kirk and 
Raven, who write : ''The great majority fail to recognize this 
truth, which is 'common' — that is, both valid for all things 
and accessible for all men, if only they use their observation 
and their understanding and do not fabricate a private and 
deceptive intelligence."^ The accidental fact that the Greek 
word for "common," current in HeracHtus' day, and an ex- 
pression meaning "with mind" sounded so nearly alike as to 
enable him to connect them in a repeated pun, evidently struck 
HeracHtus as significant.^ And although today we are more 
skeptical of the power of word-magic, we can nevertheless 
recognize the point of connecting the two meanings. What the 
pun succeeds in stressing is the natural connection between 
thinking "with rational awareness" and allowing one's 
thoughts to be guided by "what is common" — ^that is, to be 
guided by the divine Logos which is present in all things and 
discoverable by all observers if only they will open their eyes 
and their minds to the fullest possible extent. 

These considerations enable us to define with some pre- 
cision the relation and distinction between HeracHtus' state- 
ment of method and the method which may be termed, in a 
responsible sense of the word, that of mysticism. Although 
the word "mysticism" is often used loosely and pejoratively, 
it can be given and should be given a fairly definite sense, with 
regard both to doctrine and to practice. Mysticism as a 
philosophy is the doctrine that truth, at least in its higher 
forms, can be known only by participation in the divine ; and 
the practice of mysticism is the attempt to find ways of realiz- 
ing such participation. By this definition (which reflects the 
twin central themes of the most characteristic mystical writ- 
ings) ought HeracHtus to be described as mystical, or ought 
he not? There is a limited acceptance of the principle of 
mysticism in his doctrine that right method involves the 



kindling of a fiery light of intelligence within one's soul, which 
is consubstantial with the fiery intelligence that is cosmic 
activity; or, changing the metaphor, that the only valid 
knowledge about the greatest matters is that which comes to 
one who listens to the Logos and attunes his mind to it. 
Whichever metaphor is employed, that of Light or that of the 
Word, there is at least an overtone of suggestion that we come 
to know reality not by merely knowing about it (cf. Fr. 6), 
but by becoming of its nature. One can know the fiery light of 
intelligence only by becoming a fiery light of intelligence 
himself ; one can hear the divine Word only by becoming an 
expression of It through voice and deed. 

Nevertheless it would be misleading to attach the words 
^'mysticism" and ''mystical" to the method of Heraclitus, in 
view of the many connotations that are popularly attached to 
them. If, as seems likely, the adjective ''mystical" is asso- 
ciated in many people's minds with the adjective "misty," 
connoting vagueness and intellectual looseness, then nothing 
could be more opposed to Heraclitus' method and doctrine. If 
there is any element of mysticism in Heraclitus' conception of 
the upward way towards light, it is at any rate a mysticism 
not of sleep but of waking alertness. Heraclitus has nothing in 
common with the type of mystic who thinks to achieve par- 
ticipation in the divine, and thus to find truth, by going into a 
trance. In sleep and trance there is nothing but a private 
world of dreams (Frs. 15, 16); knowledge, on the other 
hand, can be attained only by becoming vigorously awake, 
which is to say by employing the faculties of sight, hearing, 
and learning (Fr. 11) in order to know the facts about the 
world of which we are part. 

Hippolytus quotes Fr. 11 directly after Fr. 116, which de- 
clares that the hidden harmony is better than the obvious. 
Whether or not the two statements were juxtaposed in 
Heraclitus' original treatise we do not know, but they may 
well have been so, for they represent two sides of the truth, 
and it would doubtless have seemed to Heraclitus appro- 

[ 25 ] 


priately effective to let them stand in sharp contrast. Hippo- 
lytus builds a logical bridge over the antithesis by explaining 
that Heraclitus ''commends and admires what is unknown 
and invisible with respect to its power," but that he holds 
what is known to men, and not what is undiscoverable, to be 
preferable. These two clauses appear in Hippolytus' Refuta- 
tion of All Heresies between his quotations of Fr. ii6 and 
Fr. 1 1 . Perhaps they do not go far to explicate the paradox, 
but I think the general meaning becomes clear as we reflect 
upon them. Heraclitus desires to avoid the kind of misty 
mysticism to which his principle of the ''identity" (or rather 
the coalescence and interplay) of opposites might easily lead 
if it were embraced without the necessary qualifications. 
Hence he warns us that eyes, ears, and understandings are 
the best witnesses; it is preferable to make our souls dry (Fr. 
46), and the way to do this is not by falling asleep but by 
opening our avenues of perception to the testimonies of 
existence as they are offered. But at the same time we should 
recognize that any piece of evidence, or any combination of 
evidences, is but a small patch of an expanse whose vastness 
transcends the powers of human understanding, and that the 
harmonies or disharmonies that become manifest to us may 
give us a very partial and distorted view of "what is un- 
known and invisible with respect to its power." 

Just as in principle the stress on the value of perceptual 
experience is balanced and qualified by a recognition of the 
hidden harmony, or secret adjustment, that is effectively at 
work in all things, so too as a practical measure one's appeal 
to the senses must be fortified by discrimination and self- 
discipline. "Eyes and ears are bad witnesses to men having 
barbarian souls" (Fr. 13) — that is (as suggested by the 
derivation of the Greek adjective), to men who can make 
only meaningless sounds like "bar bar," and so cannot com- 
municate. Again, the mere accumulation of details is not 
enough; polymathy does not teach understanding (Fr. 6). 
Although a knowledge of many particulars is important 

[ 26 ] 


(Fr. 3), yet it serves a good purpose only so far as one's 
discovery and interpretation of them are guided by listening 
to the Logos. Moreover, the outward search must be accom- 
panied by an inner search (Frs. 8, 9), for each self is a 
microcosm that reflects, in minuscule, the essential nature of 
reality at large. This is not the closet mysticism of the sleeper, 
for one's inward discoveries are not to be cut off from one's 
discoveries of the outer world, which will always provide the 
surer basis for knowledge and interpretation. 

In no important respect is the search for truth easy, nor 
will its results be obvious. For nature conceals herself be- 
neath vague indications and dark hints (Frs. 17, 18). There 
is a hidden attunement in nature, the discovery of which is 
far more deeply rewarding than the mere observation of 
surface patterns. Everything is interwoven with everything 
else ; nothing stays fixed, and even at a given moment an event 
or situation can be seen in a number of aspects, some of them 
representing sharply antithetical and "contradictory" points 
of view. Such a changing and problematical world cannot be 
known by easy or static conceptions. There must be an ac- 
tivity and resilience of the mind, corresponding to the ever 
fluctuating character of the world it seeks to know. Heraclitus, 
by his own train of reasoning, was moving toward a principle 
that Pythagoras had taught and that Plato was later to arrive 
at on the basis of a different philosophy: that truth can be 
known only by keeping the mind in active athletic trim. 

The need for mental athleticism is especially indicated in 
Frs. 16 and 19. Truth can be found only by "expecting the 
unexpected" — which is to say, only by an intellectual risk and 
venturesomeness, a practice of living always on the verge and 
being ready for whatever vicissitudes may befall. There are no 
secure mental safety-plays, and to search for them is to seek 
comforting error instead of truth. The sleeper may enjoy his 
dreams, including dreams of security and of great things 
ahead; but the waking man, whose soul is a dancing spurt 
of flame, knows that there is no security, and that the promises 

[ 27 ] 


of things to come are as fanciful and uncertain as any of the 
other dreams in which we are tempted to put our trust. For 
the future may turn out to be so utterly different from what 
we had counted on, that it is virtually as if, in the perspective 
of our present standpoint, there were to be no future for us 
at all. Whatever we see when awake is changing — which is 
to say, there is coalescence everywhere and at every moment 
between some aspects of things dying and some other aspects 
of things being born and rising to greater power. It is only 
by looking clearly and boldly into the ever-present fact of 
universal death — the death of what is familiar and the birth of 
something alien — that we can escape the net of self-delusion. 
The only valid method of inquiry is to renounce one's illusions 
of permanence and to throw one's lot unreservedly in with the 
vagaries of the changing world. 

[ 28 ] 


{20. Everything flows and nothing abides; everything gives 
way and nothing stays fixed.) 

21. You cannot step twice into the same river, for other waters 
are continually flowing on. 

22. Cool things become warm, the warm, grows cool; the 
moist dries, the parched becomes moist. 

2^. It is in changing that things find repose. 

24. Time is a child moving counters in a game; the royal 

power is a child's. 
2j. War is both father and king of all; some he has shown 

forth as gods and others as men, some he has made slaves 

and others free. 
26. It should be understood that war is the common condi- 
tion, that strife is justice, and that all things come to pass 

through the compulsion of strife. 
(2/. Homer was wrong in saying, ''Would that strife might 

perish from, amongst gods and men/' For if that were to 

occur, then all things would cease to exist. ) 

The theme of unceasing change is a very old one in philos- 
ophy. In every experience and in our normal ways of respond- 
ing to that experience it would seem that the proposition, 
''Everything changes," represents only a half-truth, although 
a very important one. For along with its multifariously shift- 
ing, fading, and vanishing aspects the world shows also some 
indications of the stable and permanent. Men may come and 
men may go, but the truth of the multiplication table does not 
budge. Or if that example is rejected as a mere conceptual 
abstraction, one can point to certain more substantial fixities 
within the physical environment itself. Although the houses 
may all go under the hill, Mother Earth remains solidly her- 
self; although individuals and societies sprout and fall like 

[ 29 ] 


leaves, the human race somehow goes on. Or even, as a 
precarious corollary of the thermonuclear age, if we should 
at last succeed in blowing up the earth and exterminating 
mankind, presumably the stars would continue on their orbits, 
majestically unmoved by our mundane fidgetings. What ex- 
perience and imagination jointly offer to our view, it would 
seem, is the changing and the permanent in varying combina- 

The philosophy of change as represented by Heraclitus goes 
a step further however. Permanence is but a relative term, his 
philosophy declares ; and what we call permanent is simply an 
example of change in slow motion or in hidden guise. All 
structures, if you observe them patiently enough and project 
your imagination far enough, are dissolving slowly; every- 
thing, as the Greeks put it, is in process of coming-to-be and 
passing-away.^ Obviously when Heraclitus says that every- 
thing flows and gives way, he does not mean that everything 
does so at the same rate or with the same degree of outward 
apparency. To our daily life of sensation and impulse it makes 
a great deal of difference whether something lasts five seconds 
or five weeks or five billion years, but such differences do not 
affect the underlying principle that sooner or later each thing 
must come to an end. The time-spans of a lightning flash, of 
a human career, and of the course of evolution from pro- 
tozoon to man, are all finite time-spans despite their obviously 
different lengths : that is the self-evident truth on which the 
philosophy of change hinges. Moreover, even in what appears 
to be most solid and perduring there is some change going on 
all the while ; even where no evidence of change is visible to 
the eye, the presumption is that everything is undergoing 
some secret alteration in one way or another at every moment. 
To Heraclitus' concretely philosophical imagination this uni- 
versal condition is symbolically represented by the flowing 
river, in which you cannot step twice, and by fire the most 
volatile of all physical things. 

When Heraclitus says that everything is undergoing some 

[ 30 ] 


degree of alteration at every moment, he must be understood 
in his own context and not in ours. A literate person today- 
would readily admit, if he stopped to think about the matter, 
that the apparently solid and inert table on which he rests his 
elbow ''is really" a collection of rapidly moving molecules, 
and that the visible color of the table "is really" a mental and 
subjective impression brought about by vibrations of a cer- 
tain frequency striking the optical nerves. Heraclitus knew 
nothing about all this, and hence he certainly could not be 
referring to it. In saying that everything is constantly chang- 
ing, what he refers to is a phenomenon much closer to actual 
experience than the scientific doctrine of molecular move- 
ment is for most of us. The meaning can be discovered by a 
simple test. Try staring at some colorful object long enough, 
and if your perceptual memory is keen you will find that some 
change has come into the visible quality during the long 
process of staring at it. The blue of the sky is likely to become 
a somewhat duller blue after too prolonged a gazing. Similarly 
with the kinds of qualities presented by hearing, taste, and 
touch. To such observations the modern reply is likely to be 
that what has changed is not the thing itself but merely our 
perception of it. But it must be remembered that this psycho- 
physical dualism, which for the past three centuries or so 
has been an idee fixe with most of us, was not a natural and 
required starting-point for thinking in Heraclitus' day. For 
him, as for most Greek thinkers, qualities are in the main what 
they appear to be ; they are properties of things primarily, of 
minds secondarily; and the distinction between the thing 
beheld and the mind beholding it was loose and fluctuating. 
Heraclitus was one who looked at the world with the eye of 
a painter, or of a child ; and his powerful speculations about 
the nature of things were reared upon that basis. In the 
qualitative sense all things are constantly changing because 
the qualities themselves are wavering, and for Heraclitus a 
thing is nothing more than the complete set of all the qualities 
and powers that belong to and constitute it. 

[ 31 ] 


Examining the concept of qualitative change more closely 
we may discover two main ways in which it is natural to 
conceive of such change as occurring: either as a passage 
from some quality to its opposite or else as a passage from one 
stage to another of a serial order. Both these extensions of 
the idea played an important role in the development of Greek 
philosophy and science. There are some changes that do not 
readily and immediately lend themselves to either of the two 
types of schematism: for instance, the purely qualitative 
change of, say, the sky from blue to pink at sunset. Blue and 
pink are not opposites in any usual sense of the word. Nor 
do they fall into an immediately evident series in the way that 
different shades of blue, or different shades of pink and red, 
might do. To be sure, when the colors are seen in a rainbow 
or in a photometer they do then occupy definite places in the 
spectrum, and thereby in that context they possess a serial 
character; in ordinary contexts, however, they do not. The 
concept of series, in one form or another, has been of the 
greatest importance in the development of science — a develop- 
ment that finds its starting-point, so far as historical evidence 
reveals, in the early cosmology of Anaximenes. Heraclitus 
gives recognition to the serial concept in his various refer- 
ences to the upward and downward ways — in Frs. 32, 34, 
46-49, and more lightly in several other Fragments. But on 
the whole his conception of change is governed by the first of 
the two schemes. Change is an alteration from opposite to 
opposite: whereas it was cool it is now warm, and whereas 
it was humid it is now dry (Fr. 22) . There is a sense, indeed, 
in which such a way of conceiving change is inescapable : a 
color change can be considered as an alteration from lighter 
to darker or the reverse, or from prettier to uglier or the 
reverse ; the passage from winter to summer can be described 
as the cold turning into warmth ; and even locomotion, when 
given a personal reference, can be regarded in terms of some- 
thing losing its hereness and taking on the opposite character 
of thereness. Such ways of thinking and speaking seem 

[ 32 ] 


strange to us because we have largely lost the radically hu- 
manistic perspective, which the Greeks still retained, and 
wherein the first step in wisdom (not the last, but always 
the first !) is to describe each thing not as we think we know it 
to be, but as it directly appears to an actively percipient mind. 
Nevertheless, even if we are concerned with change in its 
directly perceived character — which is to say, as an alteration 
of quality, not as a conceptualized movement of molecules — 
it is still an odd thing for us, and a violation of our linguistic 
and intellectual habits, to speak of the warm becoming cool, 
the moist becoming dry, and so on. We can see what the 
trouble is by a very simple mental and linguistic experiment. 
Instead of saying, "The warm becomes cool" or 'The warm 
becomes the cool," let us try the experiment of saying, **What 
was warm becomes cool." Presto, the oddity and the sense 
of paradox have disappeared. We are no longer burdened 
with the troublesome idea of something turning into its 
own opposite ; we have substituted for that paradox the more 
manageable idea of an unspecified something, a "what," 
which can successively wear the attributes of warm and cool 
in somewhat the same way in which a person might succes- 
sively wear different suits of clothes. In forming a conception 
of change we find ourselves constrained to think, as Aristotle 
has demonstrated, not dyadically in terms of two opposites 
alone, but triadically in terms of the pair of opposites and a 
substance or substratum or subject or thing in which the 
opposites are conceived successively to inhere. "It is hard to 
conceive," Aristotle remarks, "how density and rarity, for 
instance, each retaining its essential nature, could in any way 
act upon each other."^ Consequently, he concludes, "we must 
postulate the existence of a third something" that is logically 
distinct from the pair of opposite qualities which successively 
inhere in it. Sometimes the third something has a name de- 
noting a set of recognizable qualities other than the opposites 
in question. Thus, when hot soup cools or when thin soup 
thickens — omitting all reference to physical theories about 

[ 33 ] 


heat and evaporation, which are known to the average person 
by hearsay, or at best by indirect experiment and ratiocina- 
tion — we fall quite spontaneously into the triadic way of 
thinking, for the meaning of ''soup" is familiar to us through 
other properties, and we can easily think of hot and cold, thin 
and thick, as qualities that are distinct from the soup itself. 
But there are other situations, other forms of experience, 
where there is no specific subject already within purview 
and we have to invent one by a linguistic maneuver. Thus, 
when in ordinary conversation someone remarks, "It was 
cool, but now it has become warm," what do we take the 
word ''it" to refer to? If challenged on the point a person 
might perhaps retort, "the weather" ; but obviously the new 
word tells nothing more than the word "it" had done. Neither 
"it" nor "the weather" gives us any knowledge that would 
not have been already contained in the Heraclitean state- 
ment, "Cool has become warm," or "The cool has become 
the warm." The difference is not empirical, it is syntactical 
and epistemological. The preferred linguistic convention ex- 
presses a conceptual need. Most persons are uneasy in con- 
templating change in so radical a manner as Heraclitus does ; 
they require, as a conceptual prop, the idea of "something, I 
know not what" underlying the changing particulars. In this 
respect, although perhaps without knowing it, they are 

Heraclitus, on the contrary, is not an Aristotelian : neither 
grammatically nor conceptually does he share Aristotle's need 
for "a third something that endures" in any alteration from 
opposite to opposite. To him every change is a knock-down 
battle between two ontological opposites, and there is no 
referee — neither a Platonic higher Form nor an Aristotelian 
"underlying substance" — that can be regarded as standing 
logically outside the process. Even deity is no exception. Par- 
ticular gods and demigods, although superior to men (Frs. 
104, 105), are finite and will eventually be transformed from 
gods into something else (Fr. 66). On the other hand, a 

[ 34 ] 


divine name when employed in a comprehensive and absolute 
sense stands for the total self-organizing and self-destroying 
process itself; as when Heraclitus is said to declare (although 
we do not possess a direct quotation on this point) that war 
and Zeus are the same thing.^ Since conflict is the ultimate 
condition of everything, it alone merits the epithets of di- 
vinity. What is ultimate must be conceived as strife, war, and 
tension ; peace and stability are either strife in slow motion or 
at most a temporary lull between one flare-up of strife and 

If strife is fundamental, then it follows that humanly sig- 
nificant results come about not from any planning but by 
chance. Conditions may at one time be propitious to life, or 
even to the higher forms of life and thought ; at another time 
they may be destructive and annihilating. When conscious 
life beholds a set of conditions that have been favorable to its 
development, it tends to regard such conditions as produced 
by a cosmic purpose that is somehow concerned with the 
welfare and destiny of such living forms as its own. Heraclitus 
rejects any such assumption as unwarranted. If we are going 
to employ a human analogy in speaking of the guiding of the 
universe as a whole, let us describe it not as a mature, wise, 
and kindly God, but rather as an irresponsible child idly 
moving counters in a game (Fr. 24) . 

But if chance is characteristic of universal process, what is 
to be made of the statement, which Stobaeus ascribes to Hera- 
clitus, that all things occur as they are destined? To be sure, 
Diels omits the fragment from his list, but Bywater accepts 
it, and my point is that even if it is authentic — even if, indeed, 
the ensuing remark in Stobaeus that ''the destiny has the 
character of necessity" is to be taken as part of the quota- 
tion from Heraclitus,* — still there is no real difficulty about 
reconciling it with the doctrine that things happen by chance. 
The ideas of chance and necessity are not mutually contra- 
dictory but, as Aristotle has shown, ^ they represent the non- 
human aspect of things in two different perspectives. From 

[ 25 ] 


the humanistic point of view, whatever Hes outside the range 
of human planning can be said to happen ''by chance" so far 
as human purposes are concerned ; and the Greeks expressed 
the unhumanistic nature of such events by applying to them 
the metallic word dvdyKrj, ''necessity." A necessary event 
is so described not because some cosmic tyrant pushes it into 
being, but precisely because no one does so. Therefore, from 
the standpoint of anyone whom it affects, the event may be 
described as fortuitous, accidental, something that has hap- 
pened "by chance." To say that the universe flows along as 
it is destined, or by necessity, and to say that "the royal power 
is a child's," or that the counters are moved arbitrarily and 
by chance, are different ways of asserting that the major 
occurrences in the universe lie outside the range and power 
of any man or god ; they run along by themselves, as a result 
of many forces, which are most characteristically in conflict, 
but which sometimes enter into temporary and limited 
alliances, and which somehow manage in their fluctuations to 
reveal glimpses of a subtle and largely hidden harmony. 

[ 36 



28. There is exchange of all things for fire and of fire for all 
things, as there is of wares for gold and of gold for wares. 
^ 2p. This universe, which is the same for all, has not been made 
by any god or man, but it always has been, is, and will be 
— an ever-living fire, kindling itself by regular measures 
and going out by regular measures. 

JO. [The phases of fire are] craving and satiety. 

J J. It throws apart and then brings together again; it ad- 
vances and retires. 

J 2. The transformations of fire are: first, sea; and of sea, half 
becomes earth, and half the lightning- flash. 

jj. When earth has melted into sea, the resultant amount is 
the same as there had been before the sea became hardened 
into earth. 

S4. Fire lives in the death of earth, air in the death of fire, 
water in the death of air, and earth in the death of water. 

J5. The thunderbolt pilots all things. 

j6. The sun is new each day. 

J/. The sun is the breadth of a man's foot. 

j8. If there were no sun, the other stars would not suffice to 
prevent its being night. 

jp. The boundary line of evening and morning is the Bear; 
and opposite the Bear is the boundary of bright Zeus. 

40. The fairest universe is but a heap of rubbish piled up at 

41. Every beast is driven to pasture by a blow. 

Heraclitus' principle of a continually changing universe — 
what Spengler calls the ''first formulation" of his theory^ — is 
symbolized not only by such diverse figures as the flowing 
river and the child idly moving counters in a game, but also 

[ 37 ] 


and most centrally, by fire. Fire is the most significant of the 
several symbols by which the Heraclitean idea of change is 
expressed : first, because it has not just one basis of symbolic 
connection with the central idea, but at least three ; and sec- 
ondly, because Heraclitus regards fire as playing not only a 
symbolic but also a literal role of importance in his doctrine 
of the cosmos. The three main properties of fire that give it 
so important a symbolic role are its light and brightness, its 
warmth and consequent ability to effect changes such as in 
cooking, and its unique agility and power of rapid self-in- 
crease. The clear glow of wisdom in which outlines and 
distinctions are made visible, the enthusiasm and urgency to 
create, and the quickness and alertness of mind and spirit, 
which these properties respectively tend to symbolize, are 
what have given to fire (in its broadest sense, including the sun 
and the sparkle of daylight in the upper air) its perennial 
symbolic importance. These meanings are present in Hera- 
clitus' thinking too, but as is quite generally the case with 
ancient thinkers he did not distinguish between vehicle and 
tenor sharply, and it is evident that he thinks of fire not only 
as meaning something other than itself, but also as a physical 
thing (albeit a fast moving and elusive thing) playing a 
definite role in the natural world. Spengler's often brilliant 
study of Heraclitus is sometimes marred by his overemphasis 
upon the symbolic role of fire, as connoting the idea of pure 
occurrence. Spengler is right in declaring that for Heraclitus 
the most basic ontological fact is the ongoingness of things, 
and their unceasing alteration from one manifestation to 
another. But so far as he indulges his own zest for rational 
clarity by clearly formulating the idea of pure change, and 
then supposing that Heraclitus employed the image of fire as 
a symbol for this pure idea — so far as he does this, he is over- 
simplifying. Any pure scientific and philosophical idea such as 
that of abstract change is of later development, and it is an 
anachronism to attribute it to Heraclitus. Philosophical ideas 
as they were formulated by the thinkers of the sixth century 

[ 38 ] 


B.C. — even by the greatest of them — were rather Hke those 
statues of Rodin in which some part of a hving figure, usually 
the head and torso, is carved into clear and significant shape, 
but where the shaped portions seem to be a living and in- 
complete outgrowth from the mass of unhewn stone below. 
Heraclitus, like the Milesian philosophers before him, was 
making a great contribution to the clarification of physical 
process ; but the intellectual sculpturing was not yet complete, 
and a modern critic errs and distorts if he insists upon finding 
clarity to a greater degree than it exists. A historian of ideas 
must develop his ''negative capability," picking up the traces 
of form where he finds them but without pretending that the 
formal elements of an ancient doctrine can be set forth in 
totally clear outline according to present-day categories. 

As opposed to those who overstress the symbolic role of 
fire in Heraclitus, there are scholars such as Teichmiiller who 
treat fire as merely a physical actuality, ignoring or denying 
or belittling its symbolic role.^ According to this standpoint, 
Heraclitus is doing in the main the same thing that the 
Milesian physicists had done before him — seeking in one of 
the recognized elements of nature (traditionally taken to be 
fire, air, water, and earth) for a basic cause and first principle 
upon which a structure of physical science could be built. 
Teichmiiller argues (very reasonably, so far as the physical 
aspect of fire is concerned) that Heraclitus' ground for choos- 
ing this substance in preference to the water of Thales or the 
air of Anaximenes was his perception that it is the purest 
and noblest substance, having its natural residence in the 
upper sky, where there is no moisture, and where the sun can 
be a clear unadulterated fiery light. 

The emphasis upon a physical interpretation of the Hera- 
clitean fire stems from ancient times. So far as our extant 
documentary evidence can tell, it was Aristotle and some of 
his commentators, along with several writers in the doxo- 
graphical tradition stemming from Theophrastus, who gave 
ancient currency to the idea. Thus in De Caelo Aristotle 

[ 39 ] 


Speaks of those who explain the universe by reference to a 
single basic element — a remark which he specifies by adding 
that some take the basic element to be water, others air, others 
fire, and others something midway between water and air.® 
Although Heraclitus is not here mentioned by name, the 
reference is clearly to him and his followers, and the main 
thought to be noticed is that Aristotle is taking the theory of 
fire as parallel to the theories of water and air. Similarly in 
the Metaphysics, after speaking of the view of Anaximenes 
that air is "most truly the first principle (dpxri) of all simple 
bodies," he mentions by way of contrast the view of Hera- 
clitus that such a role is played by fire.* Quite naturally 
Aristotle's commentators usually fall into line with this inter- 
pretation. Asclepius contrasts Heraclitus' doctrine of fire as 
the first principle with Thales' doctrine of water and Anax- 
imander's doctrine of air. All three of the elements he takes 
as "material causes" of things. The commentators Alexander 
Aphrodisiensis and Simplicius say virtually the same thing. ^ 
Now, as has been remarked in the Introduction, in inter- 
preting Heraclitus it is necessary to bracket off our modern 
ways of thinking and enter into his own, and one of the 
respects in which this can be done is by a readiness to think 
in terms of both-and, not merely of either-or. Heraclitus' 
fire is at once the material element, the familiar flame that 
burns and crackles, and yet, too, it is the embodiment and 
symbol of change in general.® But even this double aspect 
does not exhaust its nature. For fire is also known as the 
inner light of intelligence and spiritual awareness — a light 
that is at the same time a warm but self-disciplined activity; 
and since Heraclitus does not think of a sharp demarcation 
between the inner world of self-knowledge and the outer 
world of nature, it follows that he thinks of fire as somehow 
endowed with intelligence. In acknowledging that charac- 
teristic we must avoid, if we would interpret Heraclitus 
faithfully, the temptation to personalize fire more completely ; 
to do that would be the way of mythology, which Heraclitus, 

[ 40 ] 


following his fellow-Ionian Xenophanes, repudiates. Never- 
theless, it is clear that he does think of fire not merely as 
a principle of transformation and as an element in the trans- 
formation (ideas that are variously expressed and intermin- 
gled in Frs. 28, 29, 32, and 34) but also as somehow di- 
recting and piloting all things. In Fr. 35 it is a special 
manifestation of fire in the sky, and a very audible one — 
the thunderclap, usually accompanied or preceded by a light- 
ning flash — that is said to ''pilot" all things. In Fr. 120, 
where the idea of piloting is expressed by a different but 
synonymous Greek verb, the guiding power is described as 
''intelligence." Comparison of Frs. 35 and 120 strongly sug- 
gests that the ideas of fire and intelligence were, to Heraclitus' 
mind, interchangeable or at any rate closely related and 
mutually coalescent. In addition, there are passages in the 
doxographers which show that there was a generally held 
later opinion that Heraclitus had made such a coupling. For 
instance, Hippolytus, in introducing Fr. 30, declares that for 
Heraclitus "fire is characterized by intelligence, and is re- 
sponsible for the management of the universe" ; while Stobaeus 
offers the intriguing statement, "Heraclitus held that the 
cosmos is generated not by time but by mind."^ 

Concerning the physical aspect of fire it is important to 
think of it in ancient, not modern terms. The tendency of 
modern physics, roughly since Galileo,^ has been to reen- 
visage all kinds of change as caused by, and as epiphenomena 
of, movements in space (what Aristotle calls "locomotions"), 
and to regard the moving parts of a thing, which we cannot 
see, as more real than the perceptible and tangible character- 
istics. This was not the Greek way of viewing the matter, 
and it was not Heraclitus' way. When he speaks of a process 
of continual change, he refers to an observable process of 
changing qualities of the thing as a whole, and not to the 
changing locations of its minuscule parts. However, it must 
be admitted that ancient Greece, too, had its speculative re- 
ductionists, although their characteristic ways of reducing 

[ 41 ] 


were different from ours. The so-called atomism of Leucip- 
pus and Democritus (roughly what we mean by the molecu- 
lar theory, if the molecules could be thought of as not being 
further divisible into atomic parts) was but one speculative 
theory among others. A hint of another and more curious 
sort of theory is furnished by Simplicius' remark that fire 
as Heraclitus conceived it is irreducibly fire and is "not com- 
posed of pyramids."^ Evidently there must have been a theory 
abroad in ancient Greece that fire did consist, or might con- 
sist, of pyramids. Aristotle, who regards the theory as rather 
quaint and implausible, mentions two reasons for it. On the 
one hand the pyramid is the most piercing of solid figures, 
as fire is the most piercing of physical elements — a crude 
argument, Aristotle observes. On the other hand, the pyra- 
mid (evidently he means one with a triangular base, which 
is to say a tetrahedron) is the simplest of solid figures, and 
therefore by ancient logic could be regarded as an element 
in all other solid figures. Simplicius, in commenting upon 
this passage in Aristotle, declares firmly that Heraclitus re- 
garded fire as the basic, irreducible element in all things and 
did not think of it as composed of pyramids — nor, it may 
be added, as composed of any other geometrical form nor 
of any set of locomotions. 

Fire, then — ignoring for the moment its further symbolic 
meanings — is fire as seen and felt, the familiar qualitative 
entity with which everyone is acquainted; but it must be 
added that the brightness, warmth, and activity of fire are 
both outward and inward at once, for there is not yet any 
clear division between chemistry and psychology. Regarding 
fire in this way (and not as "composed of pyramids") we 
cannot, of course, conceive of change as molecular or atomic 

How, then, is change to be conceived? The end of Fr. 29 
gives the most natural answer in terms of the dominant 
imagery : the cosmic fire becomes "kindled" and "extin- 
guished" — both processes taking place "by regular meas- 

[ 42 ] 


ures." Evidently this is a more definitely physical way of 
conceiving what Heraclitus calls in Fr. 108 the upward and 
downward ways. The double process, on its physical side, 
is described more specifically in Fr. 32. Fire is here said 
to transform itself into two other types of manifestation — 
sea and earth, representing with fire the three main stages 
of physical transformation. But in the return passage up- 
wards the earth turns again into sea, and the sea, or part 
of it, turns into fire, which is now more dramatically and 
concretely spoken of as the lightning flash (7rpr)a-Tr}p). 

It is impossible to find an English word that accurately 
translates Trpiqo-Tiqp, for the plain reason that the meteorologi- 
cal phenomenon to which the Greek word refers is not found 
in English-speaking countries. There has been a good deal 
of dispute, among both ancient and modern writers, as to 
just what the phenomenon is. Epicurus, in a passage in his 
letter to Pythocles, seems to describe it as a kind of whirl- 
wind or cyclone accompanied by a water-spout. Later, Lu- 
cretius speaks in his poem of ''what the Greeks have called 
7rpT7o-TT7p€9," and describes them as ''bursting down from 
above into the sea." In other words, as nearly as I can in- 
terpret these two passages, Epicurus thinks of the phenom- 
enon as mainly involving an upward motion ("water-spout") 
and Lucretius as mainly involving a downward. Neither of 
the passages directly mentions its fiery character; but Cyril 
Bailey, in commenting upon the latter passage, interprets 
the Greek word as implying the presence of fire. Whether 
his etymology is accepted or not, there is a confirmatory 
statement by Seneca, who, writing on types of wind, says 
that what the Greeks called Trprjorrrip is something that bursts 
into flame (inflammatur) and is a fiery whirl (igneus turbo). 
Moreover, some kind of ignition is evidently implied by 
Aetius' statement that the phenomenon is caused "by the 
kindling and extinction of clouds. "^° Consequently, although 
there is some doubt as to precise descriptive details, one 
main fact about the phenomenon is clear : that it is a mete- 

[ 43 ] 


orological occurrence which in some manner combines water, 
wind, and fire in its composition, and hence it serves Hera- 
cHtus as a vivid and appropriate exhibition of the physical 
interchange that goes on naturally between water and fire, 
although more prominently at some times and places than 
at others. 

One further remark by Lucretius may throw an incidental 
light upon Heraclitus' view of how the elements interrelate. 
He states that a Trpr^crTrip rarely spends its force over the 
land, where the hills would tend to break it up, but more 
often over the sea *Vith its wide prospect and open sky." 
This remark may be compared with Heraclitus' view of the 
sea as playing the role of intermediary between the passive 
earth and the violently active TrpiqcrTrjp. 

But how is the lightning flash of Fr. 32 related to the 
thunderbolt of Fr. 35? They are certainly not to be distin- 
guished as the ideas of lightning and thunder are distin- 
guished in our modern languages, which take the one as 
visible, the other as audible. From the various instances of 
both words in Greek literature it is clear that both a visual 
and an auditory meaning are attached to each of them. Per- 
haps the first word puts a slightly greater emphasis upon 
the visual properties while the second puts a slightly greater 
emphasis upon the auditory properties, but the two sets of 
properties cannot be separated; for, whichever word was 
employed in a given case, the Greeks tended to think in 
terms of the whole phenomenon, visual and auditory aspects 
combined. Thus Kirk is probably right in remarking that 
the thunderbolt (Fr. 35) ''may stand as a name for fire in 
general, or perhaps for celestial fire in particular. "^^ Both 
words furnish instances of ontological synecdoche; each of 
them offers a particular sort of observable phenomenon, at 
once fiery and dramatically impressive (probably tonal too), 
as an ''eminent instance" (to use Goethe's phrase) of the 
universal fiery quasi-intelligence that steers all things through 
all things (Fr. 120). 

[ 44 ] 


Now the ideas of kindling and extinguishing, as employed 
in modern Western language and thought, apply only to 
the uppermost phase of the cosmic process — the passage of 
other forms of matter into and out of fire. While it is not 
certain whether the words for ''kindling" and "extinguish- 
ing," as quoted by Clement in Fr. 29, were employed by 
Heraclitus for the lower, subaqueous phases of the process, 
the fact remains that, whatever word or words he may have 
used, Heraclitus evidently thought of the upper and lower 
phases of the process as mutually continuous and hence as 
graspable in terms of a single concept. In Fr. 33 he speaks 
of earth as "melting" into sea, and of sea as "hardening" 
into earth. In several doxographical references to Heracli- 
tus, on the other hand, it is not "melting" but rather "evapo- 
ration" that describes the passage from earth into something 
more fluid. Even if the Greek word for "evaporation" which 
is found in later writers {ava6vixiacn<s) is of later coinage, 
some idea of the evaporative process must of course have 
been familiar at a much earlier period. There is a passage 
in Aetius which suggests that it may have been Thales who 
first put scientific stress upon the idea of evaporation in nat- 
ural process. Plutarch, in his version of the same passage, is 
evidently speaking with careful accuracy, for he numbers 
his points. The third doctrine that he here attributes to 
Thales is "that even the very fire of the sun and stars, and 
indeed the cosmos itself, is nourished by evaporation of the 
waters. "^^ I can see no reason for doubting Plutarch's state- 
ment that Thales held such a view; for surely it is reason- 
able enough that Thales, starting with the postulate that the 
basic substance is water, should have conceived of evapo- 
ration as the primary process of nature. If he did so, then 
the idea of evaporation may well have been carried on in 
the Milesian school as an important scientific concept. The 
sparseness of documentary evidence prevents our knowing 
what word or words may have been employed to express 
the concept, but I would think it probable that some word 

[45 ] 


was in use — although perhaps not the "efflux" that was ques- 
tionably attributed to Anaximenes^^ — a word broad enough, 
or semantically flexible enough, to combine into one concept 
the ideas of melting, evaporation, and perhaps even of burst- 
ing into flame. It may be, too, that in referring to the con- 
trary physical process, from fire downward to earth, the 
early Milesian scientists employed a similarly broad and flex- 
ible concept to include the ideas of self-extinguishing (ap- 
plied to fire as it becomes transformed into something 
grosser), liquefying (from an airy to an aqueous state), and 
solidifying. The hypothesis is doubtless speculative, but it 
is not without some degree of scattered evidence. Now if 
the ideas of evaporation and condensation, in the broad sense 
here indicated, were current in sixth century Miletus, it is 
not at all improbable that by the end of the century Hera- 
clitus, living some twenty-five miles away, would have be- 
come acquainted with them ; nor is it improbable that, despite 
his vaunted independence of other thinkers, he might have 
been influenced by them. Even the most original of thinkers 
reaches out sometimes for other men's conceptual structures, 
which he may then adapt to his own ideas ; and it may well 
be that Heraclitus was more indebted to certain Milesian 
ways of thinking than he realized. (It is to be noted that, 
although he hurls some scathing words against Homer, He- 
siod, and Pythagoras, there is no record of his having made 
any derogatory allusion to the scientists of the neighboring 
city of Miletus.) The notion which he seems most likely to 
have taken from the Milesian school of thought, and then 
to have adapted and varied according to the demands of his 
own more active imagination, is that of the double physical 
process which might be loosely indicated by the words rare- 
faction and condensation. Since Heraclitus was above all a 
perspectival thinker, always ready and alert to readapt his 
viewpoint to the changing character of whatever might con- 
front him, and to the variously seen many-sidedness of every 
phenomenon, he would have recognized as different guises 

[ 46 ] 


of much the same thing the upward way, the process of 
thinning, the two-phased process of evaporation and con- 
flagration, and (inwardly) the struggle toward intellectual 
integrity and self-knowledge. 

But now comes a question over which there has been much 
scholarly dispute. Do the upward and downward processes 
of nature involve three main stages or four? Fr. 32 indi- 
cates the former alternative, Fr. 34 the latter. Is air, which 
is mentioned in Fr. 34 and not in Fr. 32, a real stage in 
the Heraclitean cosmology, or is it not? Certain scholars 
have gone so far as to deny the authenticity of Fr. 34, or 
to regard it as a mangled version of what Heraclitus really 
said.^* Now it is possible, to be sure, that Maximus of Tyre, 
who quotes the Fragment as it stands and who is a genial 
but not always exact thinker, may have taken from Hera- 
clitus simply the general idea of the physical elements living 
and dying in relation to one another, and in formulating it 
may have spoken loosely in terms of the more familiar doc- 
trine of the elements as fourfold, ignoring Heraclitus' three- 
fold scheme as being irrelevant to his purpose as a moralizing 
essayist. But on the other hand there is the somewhat con- 
firmatory version of the quotation as given by Plutarch: 
"The death of fire is the birth of air, the death of air is the 
birth of water. "^^ Although earth is ignored in the statement, 
the three elements that are mentioned include air, which is 
the distinctive element in Fr. 34 and is omitted from Fr. 32. 

Assuming, then, that the Fragment is quoted correctly, a 
possible explanation of the discrepancy might be that Hera- 
clitus held the two views at different stages of his philosophi- 
cal career — i.e., that he began by accepting the conventional 
notion of the four elements (fire, air, water, earth) and that 
he later simplified this conventional schema by leaving out 
air. A reason for such simplification might be suggested by 
Fr. 36, for in declaring that a new sun is born and dies 
each day Heraclitus meant, as Galenus explains, that the sun 
is moulded each morning out of the waters surrounding the 

[ 47 ] 


earth and becomes one with the waters again when it drops 
back into them in the evening/^ The passage from water to 
fire and from fire to water is here conceived as direct, with- 
out the intermediate state of air having to be assumed. 

Another possible hypothesis is that Herachtus may have 
held the threefold and the fourfold principles simultaneously, 
identifying air with soul and hence with what would later 
be called potentiality, while conceiving the three other ele- 
ments as actualities. (I grant that Heraclitus possessed no 
proper words for this Aristotelian pair of ideas, but it is 
a mark of his genius that he is repeatedly reaching beyond 
the available vocabulary in an effort to fashion and grasp 
ideas that he can only haltingly express.) Although such 
a hypothesis interprets Heraclitus as giving greater prom- 
inence to air in his cosmology than he is usually supposed 
to do, it should not be dismissed without considering one 
small piece of evidence in its favor. Sextus Empiricus twice 
mentions an opinion, held by Aenesidemus and others, that 
Heraclitus took the basic existent to be air. In one passage, 
speaking of theories about "the primary and fundamental 
elements," he says that according to some interpretations 
Heraclitus supposed such elements to be of the nature of air, 
but that according to others he supposed them to be fire. 
In the other passage, speaking of ''the existent" (to 6v), 
he cites Aenesidemus as saying that Heraclitus considered 
it to be air.^^ Nothing further is said on the subject, and it 
must be admitted that no firm conclusion can be drawn from 
such sparse evidence. Nevertheless, considering that in early 
Greek times the soul is closely associated with air, and that 
for Heraclitus soul is the first principle (Fr. 43), considering 
also the problem of reconciling Frs. 32 and 34, and consid- 
ering finally that while soul is superior to water (Frs. 44, 
49) yet from the general evidence of the Fragments of 
Chapter iv it is not ordinarily equivalent to pure fire — I 
would think that at least the hypothesis might be entertained 
that Heraclitus may have regarded the soul as somehow 

[ 48 ] 


hovering between the condition of water and that of fire, 
and therefore as being (according to the natural context of 
early Greek ontology) something rather like air. Such airi- 
ness would be conceived as an unstable and potential state, 
from which soul could either slip downward into mud or 
strive upward to the condition of fire. 

Two Fragments of the present group, Frs. 30 and 33, 
suggest dim analogies to certain more developed modern 
philosophies. In Fr. 33, which can be discussed the more 
briefly of the two, Heraclitus appears to be groping for a 
way of stating something like the law of physical conserva- 
tion. Although his apprehension of the law is of course rela- 
tively primitive and vague as compared with the exact for- 
mulation it has received in modern physics, nevertheless there 
is evidently in the Fragment a new insight, an insistence 
that in the complicated process of physical transformations 
there is nothing quantitatively either gained or lost; and 
this represents an important step in the development of sci- 
entific thought. 

Fr. 30 ('The phases of fire are craving and satiety") is 
of interest as representing what may be called the empathetic 
trend in philosophy — the tendency to interpret the essence of 
outer things and activities in the light of characteristics that 
we inwardly discover as belonging to ourselves. Physical 
events, when looked at in anthropocentric perspective, appear 
to be motivated in a manner somehow similar to our own 
behavior. But in what terms is that similarity to be ex- 
pressed? Nietzsche, Schopenhauer, and von Hartmann have 
spoken in various ways of the will in nature; Bergson of 
the elan vital that is at the heart of all activity ; others have 
spoken of mind in nature, and the like. But a single word is 
likely to say too little, or too much, or both too little and 
too much in different ways. The pair of terms, craving and 
satiety, gives appropriate contour to the scarcely sayable no- 
tion. Motive power as we know it in nature (storms, plant 
growth, and animal instincts, as distinguished from the me- 

[ 49 ] 


chanical activities of man-made machines) is not homogene- 
ous ; it does not express itself equally at all times ; there are 
periods of craving, need, and struggle, and there are periods 
of satiety and rest. The opposites are of course polar, tem- 
porary, and mutually relative, but they are nonetheless real 
and effective aspects of nature as it can be observed. 

One of the most disputed questions concerning Heraclitus' 
cosmology has to do with whether or not he believed in 
world-cycles, and more specifically in the doctrine of €/c- 
7rvp&)o-t9. This word, as used by Stoic writers, designates 
the dissolution of the universe by fire. The Stoic doctrine 
goes on to say that such dissolution is periodic, occurring 
after very long intervals of time, and being eventually fol- 
lowed by the gradual emergence of a new universe out of 
the fiery mass. The long period of time between one confla- 
gration and the next, or perhaps between the first appearance 
of things out of the cosmic fire and their ultimate extinction 
by reabsorption into it, was identified by Stoics with the 
ancient doctrine, probably of Chaldaean origin, of the Great 
Year. Among the Chaldaeans, who were devoted astrono- 
mers and astrologers, the length of such a cosmic period 
was believed to be determined by the time it took the seven 
known planets (including sun and moon) to come again 
into conjunction. In the Chaldaean tradition the length of 
the Great Year was taken as equivalent to 36,000 ordinary 
years — a figure evidently chosen on some ground other than 
exact astronomical computation. Traces of a vaguely similar 
doctrine can be found in the literatures of ancient India and 
Iran, as well as among some of the cryptic records of the 
ancient Mayas and Aztecs. In the Chaldaean doctrine it was 
further held that when the conjunction of the planets takes 
place in Cancer all things are reduced to water, and that 
when it occurs in Capricorn all things become one with, or 
are consumed by, fire.^^ Among the Greeks, on the other 
hand, the period of the Great Year was usually estimated 
as lasting either 18,000 or 10,800 years. The latter figure 

[ 50 ] 


held a special appeal to a people eager to find relations be- 
tween microcosm and macrocosm ; for it represents the prod- 
uct of 30 (the average span between one human generation 
and the next) multiplied by 360 (the formalized notion of 
the number of days in a year) ; hence, to the ancient mind, 
a Great Year of this length would represent a year of human 
generations, with the span between one generation and the 
next counting as a day. 

Did Heraclitus himself believe in the dissolution of the 
universe by fire? Among ancient writers, for the most part, 
it appears to have been supposed that he did. Admittedly the 
Fragments that are accepted as direct quotations from Hera- 
clitus offer no proof either way. Although the group of Frag- 
ments 28 to 34 and Fr. y2 can be interpreted as referring 
to such a cataclysm if Heraclitus' belief in it could be estab- 
lished on other grounds, they do not prove anything by them- 
selves in this respect, inasmuch as they can all be reasonably 
interpreted without reference to the doctrine. Each of them 
might perhaps be describing some aspect of the day-to-day 
behavior of fire in its relation to the ceaselessly changing 
world, and without any supposition of a time when fire will 
consume everything else completely. The question whether 
Heraclitus did or did not hold the doctrine of world confla- 
gration and world cycles cannot be answered on the evidence 
of the canonical Fragments alone ; most of the evidence, pro 
and con, must be circumstantial and based on indirect testi- 

The strongest argument against the supposition that Her- 
aclitus believed in such a doctrine has been drawn up by 
Kirk, on the following counts. He argues ( i ) that the entire 
tenor of Heraclitus' argument is against the doctrine, since 
''the unity of opposites upon which the Logos is founded 
depends upon the balance between them"; (2) that the doc- 
trine would contradict the emphasis upon "measures" (as 
in Fr. 29) and upon the ''exchange" that goes on between 
fire and all things (Fr. 28) ; (3) that it would contradict 

[ 51 ] 


the statement (Fr. 29) that the universal process *1s eternal 
and will never be destroyed" (Kirk's paraphrase of '^always 
has been, is, and will be") ; (4) that it would mean that 
Plato was grossly mistaken in distinguishing between Em- 
pedocles' view that cosmic unity and plurality exist in al- 
ternation and Heraclitus' view that the two states exist si- 
multaneously; (5) that ''even among Stoic sympathizers 
there were some who doubted the eWvpwo-ts-interpretation" ; 
and that the Fragments which are commonly supposed to 
support the view are of no evidential value for the purpose/^ 
The array of arguments is formidable, both by their cumu- 
lative force and because of the deserved scholarly reputation 
of their proponent. They fall short, however, of being con- 
clusive; and in order to be able to consider fairly what can 
be said on the opposite side of the question I shall briefly 
examine them one by one. 

( I ) . The first argument seems to me the strongest. Kirk 
adds : 'Tf the 'strife' which symbolizes their interaction, and 
the consequent maintenance of the tension, ceased, then the 
world would cease to be — a consequence for which Heracli- 
tus evidently rebukes Homer" (Fr. 27). Despite the self- 
refuting character of the alleged consequence, however, I 
think it can legitimately be questioned whether the conse- 
quence would logically follow in the sense that Kirk sup- 
poses. If the dominance of fire in an iKTTvp(o(TL<; were to entail 
the destruction of all strife, then admittedly a situation would 
arise — an interval of absolute peace and rest — such as is 
expressly denied by several of Heraclitus' statements. But 
would a cosmic conflagration ever be absolute? Could it, 
in Heraclitus' terms of thinking, represent an interval of 
unalloyed oneness and unchallenged stasis? The very notion 
is repugnant to Heraclitus' style of thought. But could there 
not be a periodic cosmic conflagration without any implica- 
tion of purity? There is nothing pure about the contrary 
cosmic situation — when a maximal amount of the fiery sub- 
stance has transformed itself into water and earth. Why could 

[ 52 ] 


there not be a counteractive situation, occurring at vastly 
long intervals, in which the universe somehow bursts into 
flame (as the doxographers have described the occurrence) 
with nothing more implied than that a maximal amount 
of the universal stuff (which is also process) has returned 
to a fiery condition ? Surely the cosmic fiery state would have 
to be somehow impure in order to allow the seeds of a future 
universe to emerge from it. Even if the Upward Way is 
dominant during certain cosmic periods, still the tendencies 
of the Downward Way must always be somehow latent in it. 

(2) and (3). The suggested reply to Kirk's first argu- 
ment would have some relevance to the status of his second 
and third as well. The ambiguity of the phrase "by regular 
measures" is considered in the Note to Fr. 29 (Appendix B). 
Because of that ambiguity it is impossible to know whether 
the phrase argues for or against a belief in cosmic periods. 
Moreover, to say that cosmic process always has been and 
will be, is not to deny that there may be vast periods in 
which now one set of characteristics and now another pre- 

(4). What, then, of Plato's distinction, in The Sophist, 
between Heraclitus, who is said to have declared that unity 
and plurality exist simultaneously, and Empedocles, who is 
said to have held that these opposite states occur in temporal 
succession ?^*^ Presumably unity and plurality in this context 
refer, so far as Heraclitus is concerned, to fire on the one 
hand and the universe of individuated things on the other; 
and Kirk remarks that **no supporter of an eWvpwcrt? in 
Heraclitus has been able to explain the testimony away." 
Nevertheless, one can observe that there are other places in 
the Dialogues where Plato is admittedly somewhat inaccu- 
rate and capricious in his historical references, and that there 
is some reason to suppose that his views on Heraclitus may 
have been taken from Heracliteans who were his contempo- 
raries rather than from a textual study of the older philos- 

[ 53 ] 


opher's own writings. ^^ In short, Plato's reference, although 
a stumbling block, is hardly a refutation. 

Finally, Argument 5 is not a positive argument, but con- 
sists in saying that neither the evidence from Stoic opinions 
about Heraclitus nor the evidence from the canonical Frag- 
ments is conclusive. 

On the other side of the controversy the most telling of 
the indirect evidences, in the form of testimonies that are 
not exact quotations, is found in Aristotle's De Caelo. Aris- 
totle declares : 

'That the world was generated all are agreed, but, the 
generation having occurred, some say that [the generated 
world] is eternal, others say that it is destructible like any 
other natural formation. Others again, with Empedocles of 
Acragas and Heraclitus of Ephesus, believe that there is 
alternation in the destructive process, which takes now this 
direction, now that, and continues without end."^^ 

Since Aristotle here contrasts Heraclitus' position both 
with the belief that the world will last forever and with the 
belief that it will be destroyed once and for all, it must be 
that he takes Heraclitus to believe that the world will be 
destroyed and then will be created again, in a series of cos- 
mic catastrophes and renewals, ''continuing without end." 
The catastrophic phase of such a cycle, if Heraclitus believed 
in it, would be (whatever word he himself may have em- 
ployed) what was later designated by the Stoic word €K7n5pa)o-t9. 
It is possible, of course, that Aristotle, as was suggested 
above in the case of Plato, might have erred in ascribing to 
Heraclitus opinions that were developed by certain self-styled 
Heracliteans of the fourth century. On the other hand there 
is no evidence, and I think no one has ever suggested, that 
the doctrine arose among Heracliteans of either Plato's or 
Aristotle's time. If the doctrine does not go back to Heracli- 
tus himself, the usual alternative theory has been that it 
was of Stoic origin. Yet here is Aristotle, a generation before 
the advent of Stoicism, referring to the doctrine and evi- 

[ 54 ] 


dently supposing that it dates back to Heraclitus a century 
and a half earher. Does not Aristotle's apparent supposition, 
then, offer a reasonable, although by no means conclusive, 
ground for thinking that Heraclitus may have held, or at 
least may have speculated upon, the doctrine of periodic 
world conflagration? 

There is another passage in Aristotle which indicates per- 
haps even more suggestively that he attributed the doctrine 
to Heraclitus. In the third book of the Physics he cites, per- 
haps even quotes (one cannot be sure), Heraclitus as saying 
that *'at some time all things become fire." Now if this state- 
ment were taken by itself, it could be interpreted in either 
of two ways. It could be referring ( i ) to a general confla- 
gration in which all things together enter into the fiery state, 
or (2) to the view that at different times different things 
come to the end of their individual existences and hence 
dissolve into the fire that is the basic constituent of every- 
thing. In short, it could be questioned whether *'all things" 
is to be taken collectively or dissociatively. But the answer 
is made plain, it seems to me, by the context. Aristotle has 
just laid down the proposition that ''neither fire nor any of 
the other elements can be infinite" — i.e., unlimited by the co- 
presence of other elements, other types of substance. That 
is to say, as he adds in the next sentence, "the All cannot 
either be or become any one of them." This is Aristotle's 
statement of his own view. He then adds, by way of con- 
trast, Heraclitus' view that at some time all things do be- 
come fire.^^ There would be no point in introducing Hera- 
clitus' opinion (or supposed opinion) here unless it were 
offered as opposed to the view upheld by Aristotle himself. 
Aristotle is declaring in the passage that the All, the entire 
universe, cannot possibly ever become a single substance such 
as fire, and he evidently means to cite Heraclitus as declaring 
on the contrary that at some time the All can and does become 

Although among the doxographers there are various later 

[ SS ] 


references that attribute to Heraclitus a belief in periodic 
universal conflagration, such references have little independ- 
ent value since it may well be that the information compiled 
by the doxographers stemmed, at least in large part, from in- 
formation and opinions that were current in Aristotelian 
circles. The Aristotelian commentator Simplicius, in dealing 
with the above passage from De Caelo, epitomizes what he 
takes to be Heraclitus' doctrine in the words : "Periodically 
the universe bursts into flame and periodically it becomes 
extinguished." The fullest of such statements is made by 
Aetius, as reconstructed by Diels from an epistle ascribed 
to Plutarch and from Stobaeus' Eclogues. The two versions, 
despite some minor differences, agree in attributing to Hera- 
clitus and Hippasus jointly the view that ''the universe and 
the bodies in it" are ''dissolved by fire" in the general con- 
flagration. Moreover, both versions introduce mention of the 
total fiery dissolution by the phrase, "then again" — evidently 
signifying that the general conflagration occurs not just once 
but repeatedly.^* Other doxographical testimonies to much 
the same effect could be added. Such quotations show how 
widely in later ancient times it was customary among scholars 
to impute a belief in e/cTrvpwo-t? to Heraclitus. It is a pos- 
sibility (nothing more) that those doxographers, or some of 
them, may have had the text of Heraclitus' treatise available 
to them ; but it is also possible that they may have been simply 
repeating an unverified traditional interpretation. 

In the face of these conflicting testimonies and evidences it 
is impossible to be certain today whether or not Heraclitus 
believed in a fiery periodicity of the universe. Among modern 
scholars Gomperz (the elder), Zeller, Gigon, Stock (the 
Aristotelian translator), and Mondolfo think that he did; 
Burnet, Frankel, Cherniss, Kirk, and some others think that 
he did not. The question had better be regarded as an open 
one, and fortunately its solution is not required for an under- 
standing of the essential points of Heraclitus' teaching, nor 
for an appreciation of their human significance. 

[ S6 ] 


The final Fragment of the present group (Fr. 41) serves 
as a transition to the material of the next chapter. It contains 
the only surviving statement of Heraclitus that is specifically 
about the nature of animal impulse. The anonymous Aris- 
totelian author of De Mundo quite evidently understands the 
blow in question to mean a divine blow that goads the animal 
from within,^^ rather than the blow of a whip from outside. 
(It is quite possible, to be sure, that Heraclitus, who so often 
thinks plurisignatively, may have had the latter and more 
commonplace idea in mind as well, since it could serve as an 
outer visual symbol for the former.) Although the Aris- 
totelian writer quotes the passage immediately after declaring 
that all animals are born, grow, and decay "in obedience to 
the ordinance of God," it is probable that as an Aristotelian 
he would not have had the idea of a personalized God in mind. 
If his phrase "obedience to the ordinance" is taken meta- 
phorically, his view of God as the divine goad might be simi- 
lar to the view of Heraclitus, except that Heraclitus pays more 
attention to the sudden, peculiar, and often self-warring ways 
in which the divine goad tends to manifest itself. Correspond- 
ingly, the good that is apprehended is relative (Frs. 99f¥.), 
and when it acts — whether for the weal or woe of the agent — 
it is likely to resemble the blow of a whip in its sharpness and 
unexpectedness. Such animal impulse, on the threshold of 
awareness, is intermediate between the fiery activity that is 
ever going on in the physical world and the fire of self- 
enlightenment, to which the conscious mind can rise by self- 
examination (Fr. 8) and by listening to the Logos (Fr. 2), 
which is at once other than and yet one with the individual 

[ 57 ] 


4.2. You could not discover the limits of soul, even if you 
traveled every road to do so; such is the depth of its mean- 

(43- [Soul] is the vaporisation out of which everything else 
is derived; moreover it is the least corporeal of things and 
is in ceaseless flux, for the moving world can only he known 
by what is in motion.) 

44. Souls are vaporised from what is moist. 

45. Soul has its own principle of growth. 

46. A dry soul is wisest and best. 

41. Souls take pleasure in becom^ing moist. 

48. A drunken man has to be led by a young boy^ whom he 

follows stumbling and not knowing whither he goes, for 

his soul is moist. 
4P. It is death to souls to become water, and it is death to 

water to become earth. Conversely, water comes into ex- 
istence out of earth, and souls out of water. 
^0. Even the sacred barley drink separates when it is not 

51. It is hard to fight against impulsive desire; whatever it 

wants it will buy at the cost of soul. 
5<?. It would not be better if things happened to men just as 

they wish. 
5J. Although it is better to hide our ignorance, this is hard to 

do when we relax over wine. 

54. A foolish man is a-flutter at every word. 

55. Fools, although they hear, are like the deaf; to them the 
adage applies that when present they are absent. 

5(5. Bigotry is the sacred disease. 

^y. Most people do not take heed of the things they encounter, 

nor do they grasp them even when they have learned about 

them, although they suppose they do. 

[ 58 ] 


j8. If all existing things were smoke, it is by smell that we 

would distingvtish them. 
5g. In Hades souls perceive by smelling. 
60. Corpses are more fit to be thrown out than dung. 

The question of soul is primarily the question of what it is to 
be alive — not as life is observed externally in other organisms, 
but as it is known by one who lives and is reflectively aware 
of himself as living. But while self-awareness is the most 
essential step in coming to know what life means, it must be 
accompanied by another. Men are not solipsists, and no one, 
however much of an egoist he may be, ever knows merely 
himself alone. Souls form communities, and one's growing 
awareness of what it is to be oneself is somehow bound up 
with a growing recognition of the other centers of awareness 
by which one is surrounded. Accordingly our knowledge of 
soul is at once inward and outward. The inward discovery is 
of first importance to one who seeks a sound philosophical 
method (Fr. 8), but the present group of Fragments implies 
an interplay of both procedures. 

A first task, in seeking to understand Heraclitus' view of 
soul, is to decide upon the most nearly adequate translation 
for the Greek word ^vxrj. Three English words suggest them- 
selves : soul, psyche, and self. ''Soul" has been the traditional 
translation, but the objection to it is that in many persons' 
minds the word has become entangled with certain theological 
and eschatological concepts, particularly those of divine crea- 
tion and immortality. It is even possible to hear sometimes 
the opinion expressed that the soul does not or may not exist 
— ^an opinion which can usually be found to represent the 
speaker's loose and inaccurate way of rejecting the supposed 
theological implications of the word. Now to deny or doubt 
the existence of the soul is to have lost precise contact with 
the meaning of i/zvxt? as the word is employed by the more 
reflective Greek writers who are not hampered by a dogma, 
notably Heraclitus and Aristotle. There is a sense, if one 

[ 59 ] 


narrows it down carefully, in which it is impossible to doubt 
that the thinker or doubter does in fact exist at the moment 
of doubting, and it has been a primary task of philosophical 
analysis to indicate exactly what that sense is. If "soul" could 
be taken to mean only the sense of being and being aware in 
a certain milieu of observable events, without any implicit 
postulation regarding origin, status, or destiny, the word 
would be a satisfactory one; but it has become suspect in 
many quarters today because of the associations that cling 
to it. Some modern writers, therefore, prefer to substitute the 
anglicized word ''psyche" — thinking, no doubt, that this, with 
its down-to-earth ring, is free from adventitious connotations. 
They probably delude themselves, however; for if "soul" 
carries an overtone of church and Sunday School, "psyche" 
suggests no less forcibly the psychology laboratory and the 
mental health clinic. Such associations are unintentional, but 
they are likely to saddle the thinker who uses them with 
implications that he has not fully examined. The word "self" 
is probably freer from such difficulties than either of the other 
two, although to the minds of some it might carry an in- 
dividualistic emphasis, suggesting perhaps subconsciously the 
word "selfish" as its apparent cognate. Of course, when this 
word is chosen such connotations must be avoided, for our 
language should not be allowed to rule out the possibility that 
souls may merge, partly or wholly, either with other souls or 
with a more universal and divine soulhood. The important 
thing is to choose and contextualize our words in such a way 
as not to prejudge any of the questions that may arise, and 
this requires constant vigilance. All in all, it seems to me 
wisest to translate the Greek word 4^vxrj by the word "soul" 
in the Fragments, occasionally varying it with "self" in the 
exposition that follows. 

Another point to be noted about Heraclitus' use of the word 
is that except in Fr. 48 he never employs the article with it. 
When the word occurs in the plural (Frs. 44, 47, 49) there 
is no problem of translation, for our English idiom allows us 

[ 60 ] 


to Speak of ''souls" without the article. Also in Fr. 46 there 
is no problem, for here the use of the indefinite article (which 
is nonexistent in Greek) seems to be the most idiomatic way 
of fitting up the sentence in English. In Fr. 48, which is the 
only instance where the definite article is employed, the con- 
text justifies the translation, ''his soul." But in Frs. 42, 43, 
45, and 51 the absence of the article in English may sound a 
bit queer. Translators have sometimes yielded to the tempta- 
tion of smoothing out the English by writing "the soul" in 
these instances (as Fairbanks does with Fr. 42, Lattimore 
with Frs. 42 and 51, and Freeman with Frs. 42, 45, and 51), 
but such a compromise brings into the English an air of 
definiteness and a connotation of substantiality which are not 
present in the Greek. Here it is necessary to recall the discus- 
sion in the Introduction, about certain ambiguities that are 
inherent in the language and basic conceptions of Heraclitus' 
time — in particular, the ambiguity between the noun and 
other parts of speech. "Soul," for Heraclitus, is almost a 
noun; it is more of a noun than it is anything else. Yet by 
employing it without the article he avoids a full grammatical 
commitment, and the noun in the four instances mentioned 
hovers on the brink of being an adjective, perhaps also a 
verb. The phrase, "the soul," is likely to carry, for a modern 
reader, brought up (however loosely) on Christian notions, 
a suggestion of permanence — which, of course, is absent from 
Heraclitus' conception. Soul, to Heraclitus, is quality, sub- 
stance, and activity in one. It is undoubtedly something real, 
and indeed of utmost importance (for it is only by being a 
soul, in the broad sense, that one can raise questions about 
souls or anything else) ; nevertheless we should avoid being 
trapped by accidents of grammar and idiom into saying too 
much about it. 

Since Heraclitus is building a cosmology, it becomes neces- 
sary for him to consider how soul is related to other elements 
in the fluctuating universe. In Fr. 44 souls are said to have 
their origin in what is moist, being vaporized from the 

[ 61 ] 


moisture; and in Fr. 43, on the authority of Aristotle, soul 
is said to be a process of vaporization. The word dvaOvfiiaa-L^; 
in Fr. 43 and its cognate form in Fr. 44 both connote a vapor- 
ization that is warm, or hot, or even fiery. For as remarked 
earlier, the transformation from water into air and the trans- 
formation from air into fire tended to be regarded as con- 
tinuous phases of a single process, exemplifying the Upward 
Way. Heraclitus and his contemporaries were not accustomed 
to distinguish clearly between vaporization and bursting into 
flame ; the latter phenomenon they evidently saw as the com- 
pletion and natural outcome of the former. In naively visual 
terms vapor looks like smoke, and smoke suggests fire; this 
points to the upward way. On the other hand, so far as vapor 
is moist it therein reveals its countertendency to move along 
the downward way, transforming itself into water and, if the 
tendency persists, into mud and earth. Soul, then, has its 
natural place somewhere in the area between water and fire, 
and contains within itself the possibilities of self-transforma- 
tion in either direction. That peculiar in-between state, allow- 
ing it to change itself or undergo change in either of two 
directions, is identified by Heraclitus with those stages in 
nature's up-and-down process which show themselves as 
vapor, smoke, and fiery exhalation. 

With the soul and its self-transforming power thus con- 
ceived in naturalistic perspective, it is easier to see what is 
meant by Fr. 46 on the one hand and by Frs. 47, 48, and 49 
on the other. Since soul is a dynamical something, always 
tending by a sort of inner urgency to become other than what 
it was and is, it may (if it be wise and excellent) struggle 
upwards to become drier, brighter, and more fiery, or (if 
it yields to degeneration) it may slip downwards to become 
more sodden and moist. In Aristotelian language the soul is 
a potency that can actualize itself, more or less, in either of 
the two directions. Heraclitus' vocabulary was not capable of 
expressing so logically abstract a distinction as that of Aris- 
totle between potentiality and actuality; but one of the most 

[ 62 ] 


impressive marks of his genius lies in his abiUty to reach out 
for, and darkly adumbrate, ideas that are beyond the natural 
semantic range of his somewhat primitive language. 

The situation is somewhat complicated by the fact that in 
the main doxographical passage from which Fr. 44 is taken, 
the Fragment appears as a codicil to the famous remark 
about the impossibility of stepping twice into the same river 
(Fr. 21). As reported by Arius Didymus the total passage 
runs as follows : *To those who step into the same rivers, 
other waters are continually flowing on. And souls are vapor- 
ized from what is moist." It is hard to see any clear connec- 
tion between the two sentences ; and therefore I have judged 
it best to present them separately as distinct Fragments in 
the present arrangement. Nevertheless, the words with which 
Arius Didymus introduces the double passage suggest that 
he saw a connection or was trying to see one. What he says 
is : "Wishing to make clear that the souls, as they rise up in 
vapor, become intellectually aware, he [Heraclitus] repre- 
sents them in the likeness of rivers, speaking thus . . ." ; and 
then comes the double quotation. Although this explanation 
may seem to confuse the matter more than ever, I think that 
perhaps a further clue may be found in another passage from 
Arius, in which he attributes to Heraclitus the view that soul 
is ''a perceptive exhalation."^ Evidently, by one metaphor and 
another, Heraclitus is trying to convey the idea that the soul 
comes into existence out of certain moist elements in nature 
(possibly he may have in mind the moisture of the maternal 
womb), and that the mystery of emergent selfhood, which no 
theory can ever satisfactorily explain, can be described figura- 
tively as a being exhaled or vaporized from that generative 
moisture. Aristotle, in De Anima, employs a more developed 
philosophical vocabulary to grapple with the same problem; 
but I cannot see that his characterization of the affections of 
the soul as ''meanings (Xoyot) subsisting in matter"^ comes 
any closer to the heart of the problem than Heraclitus' more 
naively picturesque representation had managed to do. 

[ ^2 ] 


But while the naturaHstic perspective furnishes an essential 
part of the truth about soul, it does not and cannot furnish 
the entire truth. Two contrasting and mutually irreducible 
aspects of soul must be considered together, if Heraclitus' doc- 
trine is to be understood. On the one hand, soul is an emergent 
and finite phenomenon of the natural world; it is born out 
of moisture, and eventually (although not necessarily at the 
moment of what we call death) it will pass away again. On 
the other hand a soul, during the span of time in which it 
is alive, possesses a real though limited autonomy. In this 
connection Fr. 45 is significant ; for to say that a soul has its 
own principle of growth is to say that it must be understood 
not as being pushed into activity from without, but as be- 
stirring itself from within — like a fire rekindling itself from 
a tiny spark. One can know what soul is only by having gone 
through such a process of self-rekindling. Soul, which is to 
say selfhood, is unique in that it alone has the double prop- 
erty of existing and of knowing its own existence. In the his- 
tory of philosophical speculations materialism has taken the 
one aspect as basic, idealism the other. But in Heraclitus' 
view neither of these opposing systems of thought is compre- 
hensive enough to express the truth. For reality, he holds, 
is paradoxical at bottom (Fr. 17) ; and that out of which 
everything else is formed (Fr. 43) can surely not be known 
by any single kind of intellectual maneuver (Fr. 42). That 
soul is a more or less accidental product of the natural world 
and that soul is somehow significantly self-determining are 
two warring but ineradicable truths about soul, which an 
awakened intellect will always hold in unresolved tension. 

The principle that the soul must strive toward the fullest 
possible activity in order to know truly, is differently repre- 
sented by the contrasting metaphors of dryness and moisture 
(Frs. 46, 47) and by the analogy of the sacred mixed drink 
(Fr. 50). The peculiar concoction which the word KVKecov 
denotes was a ceremonial beverage made of barley, grated 
cheese, and Pramnian wine, and obviously these ingredients 

[ 64 ] 


tend to separate when the drink is allowed to stand. The very 
character of the beverage can be maintained only by constant 
stirring. Theophrastus, who quotes the passage, introduces it 
with the explanation that there are certain things which hold 
together only by being in motion, and which lose their essen- 
tial nature when the motion ceases.^ Of course the general 
Heraclitean position is that all things without exception have 
their being in motion; nevertheless the principle is more 
conspicuously true of some things than of others, and it is 
especially and most significantly true of human selves. To hold 
on stubbornly to one's way of life is to lose it ; whereas to live 
always on the verge, always in readiness for whatever may 
come, ''expecting the unexpected," is to meet life on its own 
terms and thus to become one with it in the only way that 
is possible, through yielding oneself up to its law of continual 
change. An unstirred self, like an unstirred barley-drink, 
tends to decompose, breaking up into dregs of material im- 
pulse on the one hand and ghostly ideal aspirations on the 
other. The perceptions and ideas arising from such a state 
of affairs tend to become illusory. Truth about the variegated 
and paradoxical world we live in can come to us only as our 
thoughts and sensitivities are constantly entering into new 

The ethical tone of Frs. 46 to 51 stands in sharp contrast to 
the statement (Fr. 108) that the way up and the way down 
are one and the same. The present Fragments affirm that the 
upward way toward light, dryness, and intellectual awareness 
is preferable to the downward way toward moisture and 
drunken muddle. What, in view of this avowal, is meant by 
the seeming indiff erentism of the later Fragment ? The solu- 
tion must be found in the different perspectives that the two 
kinds of statement represent — the one ethical and personal, 
the other cosmological and universal. In the perspective of 
here and now, with a choice of paths before me, it is pertinent 
to judge one path better than the other; in fact, a failure to 
do so is even self-delusory, because a self that tries to avoid 

[ (>s ] 


choosing makes a choice in that very avoidance, in letting 
its action be determined by dark impulses instead of by 
lighted reason. A self in its awareness knows the difference 
between climbing and slipping, and it knows that the former 
alternative represents its fulfillment, and that the latter repre- 
sents intrinsic defeat. On the other hand, each moral struggle 
and moral choice, and the unique truth it expresses, is em- 
bedded in time as an event amidst innumerable other events. 
From the cosmic point of view the individual event, whatever 
its quality and temporal importance, is insignificant; the 
implicit attitude is symbolized by the divine child recklessly 
moving counters in a game (Fr. 24). Of all the paradoxes 
in Heraclitus' philosophy there is none more fundamental 
than this one of the simultaneous validity of the two attitudes, 
valuational and trans-valuational. 

The statements about smelling (Frs. 58, 59) require some 
remark. The obvious implication is that souls in Hades are 
composed of smoke, and that for this reason they must per- 
ceive and be perceived by smelling and being smelled.* The 
curious hypothesis is another consequence of Heraclitus' 
view, mentioned earlier, that smoke, cloud, and vapor are but 
different forms of the state of things intermediate between 
fire and water, and that soul belongs ontologically in this area. 
Being vaporous a soul is also smoky, and the question under- 
lying the two Fragments is, What kind of awareness is pos- 
sible in a disembodied, hence a smoky state? The answer 
given is that as ghosts we could no longer perceive and dis- 
criminate by sight, touch, and hearing, but only by smell. Her- 
aclitus appears to be reaching toward the idea that there are 
modes of existence essentially different from that of the famil- 
iar world which we come to know by familiar means. Perhaps 
this interpretation suggests a new and not altogether re- 
assuring twist to what is suggested by Fr. 67. Truly the 
question of what it is to be a self is a very deep mystery (Fr. 
42) ; and Heraclitus is suggesting that there may be aspects 
of the mystery, totally ungraspable in our present state of 

[ 66 ] 


being, which may become temporarily known to us, tempo- 
rarily even a part of us, in some mode of being that succeeds 
the shock of death. The main evidences of Heraclitus' view 
upon this subject are to be found in the next group of Frag- 

[ 67 ] 


6i. Human nature has no real understanding ; only the divine 
nature has it. 

(62. Man is not rational; only what encompasses him is in- 

6 J. What is divine escapes men's notice because of their in- 

64. Although intimately connected with the Logos, men keep 
setting themselves against it. 

6^. As in the nighttime a man kindles for himself (aTrrerat) 
a light, so when a living man lies down in death with his 
vision extinguished he attaches himself (aTrrerat) to the 
state of death; even as one who has been awake lies down 
with his vision extinguished and attaches himself to the 
state of sleep. 

66. Immortals become mortals, mortals become immortals; 
they live in each other's death and die in each other's life. 

6j. There await men after death such things as they neither 
expect nor have any conception of. 

68. They arise into wakefulness and become guardians of the 
living and the dead. 

6g. A man's character is his guardian divinity. 

70. Greater dooms win greater destinies. 

7 J. Justice will overtake fabricators of lies and false witnesses. 

J 2. Fire in its advance will judge and overtake all things. 

/J. How can anyone hide from that which never sets? 

(/4. When some visitors unexpectedly found Heraclitus 
warming himself by the cooking fire : Here, too, are gods.) 

75. They pray to images, much as if they should talk to 
houses; for they do not know the nature of gods and heroes. 

y6. Night-walkers, magicians, bacchantes, revellers, and par- 
ticipants in the mysteries! What are regarded as mysteries 
among men are unholy rituals. 

[ 68 ] 


77. Their processions and their phallic hymns would he dis- 
graceful exhibitions, were it not that they are done in honor 
of Dionysus. But Dionysus, in whose honor they rave and 
hold big feasts, is the same as Hades. 

J 8. When defiled they purify themselves with blood — as 
though one who had stepped into filth should wash himself 
with filth. If any of his fellowmen should perceive him 
acting in such a way, they would regard him as mad. 

7p. The Sibyl with raving mouth utters solemn, unadorned, 
unlovely words, but she reaches out over a thousand years 
with her voice because of the god in her. 

Human souls or selves exist not only in relation to the 
material substratum from which they have been vaporized; 
they exist also in significant relation, or in possibility of 
significant relation, to the divine Logos that permeates all 
things — which is to say, all activities. There is a passage in 
Sextus Empiricus which, although it does not appear to be 
meant as an exact quotation, is offered as a statement of 
Heraclitus' doctrine on this point. Sextus begins by quoting 
Euripides' address to Zeus in The Trojan Women : "To see 
into thy nature, O Zeus, is baffling to the mind. I have been 
praying to thee without knowing whether thou art necessity 
or nature or simply the intelligence of mortals."^ The divine 
reality which may be called by the name of Zeus (when the 
name is employed seriously and not with mythological flip- 
pancy) is not other than what Heraclitus symbolizes as the 
divine Logos; and the passage from Euripides is joined in 
Sextus' text with the following paraphrase of a part of Hera- 
clitus' doctrine : 

*'So it is by in-breathing the divine Logos that we become 
intelligent, according to Heraclitus. During sleep we are for- 
getful, but we become mindful again on waking up. For in 
sleep the pores of the senses are closed, so that the mind in us 
is shut off from what is akin to it in the surrounding world, 

[ 69 ] 


and its connection with outer things is preserved only at a 
vegetative level through the pores of the skin. Being thus 
cut off it loses its formative power of memory. But when we 
wake up again, it peers out through the pores of the senses, 
which serve as little windows, and by thus entering into rela- 
tion with what surrounds us it regains the power of reason."^ 

While there is a possibility that Sextus may be supplement- 
ing Heraclitus' doctrine of the superiority of waking con- 
sciousness over sleep by adding a physiological explanation 
of his own, at any rate the explanation seems to be consistent 
enough with Heraclitus' known teachings, and thus it may be 
taken as either a paraphrase or a natural development of what 
he said. Whatever the exact physiological process may be, the 
important thing for Heraclitus is that in sleep, as well as in 
mystical trances, we become deceived by dreams and hal- 
lucinations (Frs. 15, 16) ; only in waking periods, and in the 
most intensely alive of them, can we achieve some momentary 
glimpse of what truly is. 

There is a further remark which Sextus makes shortly 
after the passage just quoted ; and I add it because, once again, 
it seems to be intended as a characterization of Heraclitus' 
view, and in any case is not out of line with it. Sextus writes : 

''Just as coals when brought close to the fire undergo a 
change that renders them incandescent, while if moved away 
they become extinguished; so likewise that portion of the 
surrounding milieu that is making a sojourn in the body, in 
losing contact with the surrounding milieu, therein loses its 
rational character by the separation, for its only communion 
with the outer universe now takes place through the body's 
very numerous pores." 

The idea is that the self becomes most intensely alive and 
aware not in isolation, which can only breed illusions, but 
when it is most keenly alert to what is going on in the world 
around it — alert in such a way that like smouldering coals 

[ 70 ] 


it draws into itself more fiery substance from the enveloping 

The last of this preliminary series of statements by which 
Sextus characterizes the doctrine of Heraclitus has already 
been mentioned in Chapter i, but it deserves to be recon- 
sidered in the present context. Sextus writes : 

''So Heraclitus asserts that the common and divine Logos, 
by participation in which we become rational, is the criterion 
of truth." 

The latter part of Fr. 43, supplemented by Frs. 46 and 50, has 
now thrown further light upon the meaning of the statement. 
Participation in the Logos is not self-isolation and it is not 
stultification of the intelligence; on the contrary, since the 
Logos is the principle of continual motion and change, which 
is also symbolized by the divine fire, it can only be by making 
our souls dry and fiery, our perceptions keen, and our wills 
unsubjected to cloying emotions, that we can really participate 
in the Logos instead of feeding our fancies with mythological 
vulgarizations of it. Such real participation, through mental 
alertness to the ever-changing but objective world, is what 
provides us with the criterion of truth. 

In speaking of Heraclitus' religious philosophy, or of the 
religious dimension of his philosophy, we must avoid the 
familiar Christian associations that are likely, even in dis- 
guised form, to cling to the word. There is no supposition in 
Heraclitus' thought of a single universal God who is at once 
personal, omniscient, and (despite all appearances to the con- 
trary) deeply concerned about the ultimate destiny of man- 
kind in general and, for better or worse, of each particular 
human individual. There is, to be sure, in Heraclitus' view, a 
unity and coherence that is somehow present as a hidden other 
aspect of the plurality and diversity of things that is seen 
everywhere, and he even goes so far as to describe that cosmic 
oneness as ''wisdom" (Fr. 119) and "intelligence" (Fr. 
120). Such words, however, express a metaphor intended to 

[ 71 ] 


suggest the mysterious organizing power that pervades the 
universe as a whole, manifesting unUmited possibiHties of 
comeback after defeat and of new creation after destruction 
and death. If we interpret the metaphor Hterally, we shall 
overstress the human analogy implied in the ideas of wisdom 
and intelligence, and thus shall fall into anthropomorphism. 
Xenophanes, as remarked in the Introduction, had struck a 
new note in Greek religious philosophy by ridiculing the 
human and even racial characteristics that have variously 
been ascribed to God, by speculating that if oxen could paint 
they would represent God in the likeness of oxen,^ and by in- 
sisting that God in his true nature must transcend all such 
characteristics. No doubt Heraclitus had been influenced by 
Xenophanes' teaching in his youth ; at any rate in his mature 
philosophy he carries the principle still further. Granted, the 
surviving fragments of Xenophanes are so sparse that the 
full extent of his innovations is uncertain; but there is no 
evidence that he pursued the ethical implications of his doc- 
trine. God, he declares, is not to be pictured as snub-nosed 
in the manner of the Ethiopians, or as blue-eyed and red- 
haired in the manner of the Thracians, or as moving about 
from place to place; there is nothing said, however, about 
God being beyond good and evil. It was left for Heraclitus to 
take this final step, to which the rigorous logic of transcend- 
ence would have to lead. For whenever we ascribe ethical 
characteristics to God we inevitably appeal to idols of the 
tribe, and usually of the marketplace too. God is, we tend to 
think, on the side of men as against tigers, snakes, and deadly 
germs (evidence to the contrary notwithstanding), and even, 
at a proper time in history, on the side of democracies as 
against dictatorships. Such partisan views, in one form or 
another, tend to recommend themselves to the popular im- 
agination whenever it concerns itself briefly with religious 
matters. The grounds for them, however, become increasingly 
flimsy when one examines them philosophically without the 
rosy lenses that our fears and hopes usually contrive to set be- 

[ 72 ] 


fore our vision. To speak seriously of God's transcendence is to 
suggest that he has a nature that is independent of our images 
and conceptions of him, and to ask unflinchingly what that na- 
ture is. The way of philosophy is to do just that, and Heracli- 
tus was above all else a philosopher. If God is truly God and 
universal, then he must be the God of all that exists and moves 
and struggles in any way whatever — the God alike of gnats 
and vipers, of angels and devils, men and mice. Christians 
and cannibals, creative genius and destructive cataclysm, 
without showing any special and prolonged favoritism to any 
of them. A sort of intelligence displays itself in God's nature 
and operations, Heraclitus observes, but it is an intelligence 
only remotely and abstractly like anything we know on the 
human level, and we shall mistake it profoundly if we allow 
any further human associations to color our description of it. 
Now in judging that the divine intelligence is so utterly dif- 
ferent from our everyday intelligence, we are likening it, 
from the human partisan standpoint, to something vastly 
indifferent and irresponsible, like a child arbitrarily moving 
counters in a game (Fr. 24). This, too, is a metaphor of 
course, but it is usefully corrective of certain other and more 
familiar metaphors that tend to influence our thinking. In 
particular it is a metaphor that should be remembered when 
reading the Fragments in which Heraclitus ascribes oneness, 
intelligence, and godhead to the cosmic All. Statements about 
the universe are but half-truths — true and false at the same 
time, as Heraclitus likes to say — and the statement that in- 
telligence guides the universe must be balanced by the counter- 
statements that the thunderbolt pilots all things (Fr. 35) and 
that "the royal power is a child's" (Fr. 24). 

When the word *'gods" is used in the plural Heraclitus is 
speaking about something quite different from the ultimate 
cosmic mystery. Fr. 74, the only one requiring stage direc- 
tions, evokes a charming picture of Heraclitus, despite his 
later reputation as a gloomy and forbidding philosopher, 
meeting the situation with light grace when, on a certain occa- 

[ 73 ] 


sion, some visitors found him warming himself by the kitchen 
stove. It was not a place where a Greek gentleman would nor- 
mally have been found, and to a lesser man some explanation 
might have seemed to be called for. Heraclitus, who was not 
one to be bound by petty rules of etiquette, remarked sim- 
ply, "Here, too, are gods." The notion that gods are every- 
where, or at any rate are to be found in unexpected places, 
was not new with Heraclitus. Nearly a century before him 
Thales had declared that all things are full of gods. Aristotle, 
in quoting the remark, takes it to mean that "soul is diffused 
throughout the entire universe" ; Simplicius, commenting on 
the passage in Aristotle, takes Thales' statement to mean that 
"the gods are blended with all things," and he adds his own 
opinion that "this is strange."* The idea is strange enough at 
first sight, no doubt, but it becomes less so when it is ex- 
amined without mythological preconceptions of what the 
nature of gods must be. To be blended with gods evidently 
meant in the sixth century much the same thing as to be 
blended with soul ; for Aristotle reports that Thales reasoned 
that the magnet must be imbued with soul because it causes 
motion in the iron. Evidently, then, Thales believed that while 
the gods, or centers of soul-force, are everywhere, they are 
more active or more conspicuous in some things than in 
others. There is nothing in the view that is inconsistent with 
Heraclitus' known utterances, and quite possibly he held 
something very like it, but unfortunately there is a shortage 
of available Fragments that bear clearly upon the point. 

Two beliefs of Heraclitus are fairly evident, however: 
first, that gods are higher types of being than men are (Fr. 
104) ; secondly, that gods are nevertheless finite and mortal. 
The latter belief is clear from Fr. 66: "Immortals become 
mortals, mortals become immortals ; they live in each other's 
death and die in each other's life." The words "immortals" 
and "mortals" are traditional synonyms for gods and men 
respectively ; Heraclitus' preference for the more paradoxical 
form of expression is characteristic of him. Of course in 

[ 74 ] 


strict logic it is impossible that anything immortal should 
ever become mortal, for if it were to do so it could not have 
been truly immortal in the first place ; and vice versa. But in 
the universe as Heraclitus envisages it there is nothing truly 
immortal in the literal sense — except, indeed, the endless 
process of mortality itself. The meaning of the passage ap- 
pears to be, then, that in the whirligig of universal change 
there comes a time when a god dies and the soul-force that 
constituted him turns into something else, quite possibly into 
a man; likewise men, when they die, may turn into gods or 
something like gods for a while. This interpretation of the 
passage is confirmed by Clement of Alexandria; for the 
version that he gives of the passage is simply, "Men become 
gods, gods become men." (The verb, in both versions, is 
omitted and is left for the reader to supply.) It seems likely 
that Hippolytus' version, which is employed in Fr. 66, is the 
more authentic one — for reasons discussed at the beginning 
of Appendix B. At any rate, whatever the original verbal 
expression may have been, it seems evident enough that 
Heraclitus believed that there occur, at least in some instances, 
metamorphoses of men into gods and of gods into men. 

Now if men become gods it must be that they do so after 
the end of human life, and that therefore the soul or self must 
survive death in some way or other. Although certain scholars 
have denied that Heraclitus believed in the survival of the 
soul after death, on the ground that the soul like everything 
else must be in constant flux, I do not see how his acceptance 
of such a belief can be denied when the evidence of Frs. 59, 
65, 66, 6y, 68, and possibly 70 is fully considered. The ques- 
tion here is of survival, not of immortality. The older dis- 
tinction between the two ideas has become lost in modern 
Christendom, where the notion of a future life, whether 
aflfirmed or denied, ordinarily carries with it the idea of im- 
mortality, which is to say of ultimate deathlessness and eternal 
life. Naturally it would be inconsistent with Heraclitus' 
philosophical position to say that any soul lasts forever. 

[ 75 ] 


Nothing lasts forever, except activity. Nevertheless, even in a 
ceaselessly fluctuating universe some things last longer than 
others. In Plato's Phaedo, after it has been agreed that the 
soul probably survives the death of the body, the question 
comes up as to whether the soul might not possibly outlive 
several bodies in which it was reincarnated and yet finally 
perish, much as a man might successively wear out several 
garments and yet die in the end.^ Despite Socrates' specious 
refutation of the proposal it was a more plausible hypothesis 
to Greek thinkers than it usually is to us, and the evidence 
seems to be that Heraclitus believed in something of that 

But if Heraclitus held a theory of survival, as I believe he 
did, his indications of the nature of that survival are tan- 
talizingly incomplete in the Fragments. There is his general 
remark that the experiences awaiting a person after death 
are so utterly strange as to defy previous conception or ex- 
pectation (Fr. 67), which gives a more specific application to 
the general truth uttered in Fr. 19. If our awareness continues 
beyond death at all, it may find the ghostly state a very 
unexpected one indeed, and many an ancient philosopher has 
warned against encumbering ourselves at the threshold of 
death by clinging to any preconceived notions of what is to 
come. Fr. 68, ''They arise into wakefulness and become 
guardians of the living and the dead," is difficult to interpret 
because of the impossibility of discovering just what ''they" 
refers to. Evidently not to the dead in general, because these 
unidentified ones are to have dominion over the living and 
the dead alike. Perhaps they are the souls who in life have 
refrained from becoming moist and have made themselves 
into dry lights. 

Fragment 65 offers one of the most mysterious statements, 
and I suspect one of the most important, in all the extant 
Heraclitean corpus. Numerous variant readings and inter- 
pretations have been proposed, a few of which are mentioned 
in Appendix B. Fairbanks' well-known translation, based 

[ 76 ] 


upon Bywater's overcautiously edited version of the text, 
"Man, like a light in the night, is kindled and put out," ex- 
presses a rather banal idea and loses the gist of what Hera- 
clitus is saying. The present translation is based upon a 
restored version of the traditional text, as explained in Ap- 
pendix B ; and it indicates, I believe, darkly but in a way that 
no other Fragment does, one indispensable aspect of Hera- 
clitus' deeply paradoxical view of the nature of death. 

The most obvious fact about the Fragment is that it hinges 
upon a serious pun. The verb amj^iv in the active voice means 
both ''fasten" or "attach" and also "kindle" or "set fire to." 
Any Greek lexicon gives both meanings as a matter of course. 
In the middle voice, which is employed on at least the first 
two of the word's three appearances in the present Fragment, 
the first of these meanings would become "attach oneself to" ; 
the second would become either "strike a light for oneself" or 
"kindle oneself," "burst into flame." Now the opening clause 
of the Fragment is clear enough : as the basis of comparison 
it describes the action of a man at nightfall kindling a light 
for himself because his vision is impaired by the darkness. It 
is the next occurrence of the verb that is crucial. On the sur- 
face it makes a pun, and thus the middle part of the Fragment 
may be taken to say : "Similarly a living man, when his vision 
is extinguished by death, attaches himself to the state of 
death" (literally, to [the state of] a dead man). But if this 
were the entire meaning, the statement would be trivial and 
the pun would be a childish trick. So simple an interpreta- 
tion is unacceptable; for Heraclitus, even at his wittiest, is 
always deeply in earnest. The solution, I believe, is that the 
verb on its second occurrence carries two meanings at once : 
in addition to the meaning just given there is connoted the 
further idea of lighting oneself up, of bursting into flame. This 
second meaning, when applied to the sentence, produces the 
following translation: "Similarly a living man, when his 
[human] vision is extinguished in death, flares into flame on 
achieving the state of death." The first meaning is the princi- 

[ 77 ] 


pal one; the second is a semantic overtone, intended lightly 
and yet seriously — at once a question mark and a reminder 
that there is always another and opposite side to every situa- 
tion, and that there is always more than meets the eye in 
anything so essential and ontologically primary as the transi- 
tion known as death. The mystery of death is not something 
that can be expressed in straightforward language. All such 
language carries with it associations that have been developed 
out of the familiar experiences of everyday human life ; how, 
then, could it possibly be adequate to describe the nature of 
something so utterly unfamiliar and mysteriously trans- 
human as after-death experience, which "we can neither ex- 
pect nor have any conception of" (Fr. 67) ? Heraclitus, like 
every serious poet, perceives the limited expressiveness of 
everyday language and employs certain devices, particularly 
metaphor and paradox, for pointing, however uncertainly, 
beyond it. His use of the verb on its second occurrence in the 
Fragment, then, is plurisignative ;^ the reader's mind is chal- 
lenged to think in two directions at once — a challenge that 
must sometimes be present in all good writing. 

The final clause of the Fragment is less important, and 
looks like an afterthought, thrown in for good measure. It 
does not seem to me to clarify the strange, complex, and in- 
tentionally ambiguous point of the central clause ; but there it 
is, and we must make of it what we can. The principal ques- 
tion about it is, how to interpret the verb, which occurs for 
the third time here. Does it carry on the double meaning from 
the previous instance ? Although a sure reply is impossible, I 
would think not. Heraclitus has declared forcibly that any 
seeming illumination in the state of sleep is merely private 
dream, not real awareness (Frs. 15, 16). Granted that he 
might characteristically have in mind some contrary aspect 
of the matter, and that sleep despite its sogginess might yet 
emit sparkles of flame, nevertheless there is no warrant in the 
existent Fragments for asserting such an interpretation. On 
the basis of the fragmentary evidence of Heraclitus' teachings 

[ 78 ] 


we had better assume that the verb on its third occasion means 
simply "attach oneself" or ''become attached" and does not 
mean ''burst into flame." An incidental result of this inter- 
pretation would be to discover a neat little structure in the 
semantic variations of the verb : at first it would mean only 
lighting, then both attaching and lighting, then only attach- 
ing. It is also possible that Heraclitus intends the final 
meaning to be in the passive rather than the middle voice; 
for in the present tense, which is employed, the two voices 
would be grammatically indistinguishable. The final clause 
would then read : "even as one who has been awake becomes 
joined with the state of sleep." But it seems likely that Hera- 
clitus, having preceded the age of grammarians, would not 
explicitly have thought of the distinction between middle and 
passive where there was no difference in grammatical form, 
and that consequently the meaning in the present instance 
would straddle the distinction. 

But the foregoing interpretation of Fr. 65 leaves us with a 
conundrum: how to reconcile the notion of death that it 
appears to imply with the notion implicit in Fr. 49 ? The one 
Fragment envisages death as a flaring up in flame, the other 
identifies it with passage into a watery state. The one sees 
death in the perspective of the upward, the other in the per- 
spective of the downward way. Does Heraclitus mean that 
some souls at death go one way, some the other? This is 
evidently Kirk's answer, for he writes : 

"If, then, when the body dies the soul either becomes water 
or remains fiery, and becomes more fiery still, what is the 
factor which determines this issue? Clearly, the composition 
of the soul at the moment of death ; ... if the amount of water 
at the moment of death exceeds the amount of fire, presumably 
the soul as a whole suffers the 'death' of turning into water : 
but if the soul is predominantly 'dry,' then it escapes the 
'death' of becoming water and joins the world-mass of fire. 
This is deduction, but I think permissible deduction."^ 

[ 79 ] 


With this interpretation of Kirk's I would agree on the whole, 
but with the speculative qualification that perhaps a soul 
might undergo both processes at once. The upward and down- 
ward ways are simultaneously active in every soul, although 
to different degrees, and every soul is in some state of tension 
between upward and downward pulls. All things being in a 
state of change, souls included, it may be that the warring 
elements in a soul pull apart sharply at death, the one part 
flaring up like a light in the nighttime, the other part sinking 
coldly into a watery and earthy inertness. 

Perhaps the foregoing interpretation may throw some light 
upon the meaning of Frs. 71, 72, and 73. Since the first and 
third of them are quoted by Clement of Alexandria and the 
second by Hippolytus, there have been critics who have dis- 
counted the quotations on the ground that these Christian 
writers might have touched them up for their own evangelical 
purposes. But Clement and Hippolytus, whatever their gen- 
eral intent, were not using Heraclitus evangelistically : they 
quoted him as a pagan philosopher, and Hippolytus did so 
with the further aim of showing him to be at the root of a 
heresy that had crept into Christianity. There does not seem 
to have been much reason why they should have misquoted 
him, and moreover there is cumulative evidence that Hip- 
polytus, for one, quotes (however he may afterwards inter- 
pret) with accuracy and care. Now the quotation given by 
Hippolytus is, "Fire in its advance will judge and condemn all 
things" ; and if this can be accepted as a genuine utterance of 
Heraclitus there should be no difliculty about accepting the 
two quotations by Clement, for all three of them are in much 
the same vein. 

But of course, in order to accept the three quotations as 
authoritative it is necessary to be able to see how they fit, 
even though not tightly, into Heraclitus' otherwise known 
cosmology. Now the fundamental principle to keep in mind 
when reading any of his utterances is that everything has 
another and contrary aspect, to be seen only by a mind that is 

[ 80 ] 


active enough to be able to step into a sometimes wildly dif- 
ferent perspective from the one with which it started. There 
is no such thing as the merely physical ; that is a conceptual 
abstraction that men have developed as one of the instru- 
ments of a technological age. The Heraclitean fire, as we 
have seen, is at once physical and more than physical: it has 
a psychical aspect (as inner quickening and illumination), a 
metaphysical aspect (as eternal process), and a moral aspect. 
Now if Heraclitus believed in cosmic cycles — a possibility 
discussed in Chapter iii — then at those periods of time when 
the cosmic fire flares up into utter conflagration, there is still 
a plural significance involved. Fire, at such crises in the life 
of the cosmos, consumes all things, but it may do so in either 
of two ways. It may burn the individual to a cinder or it may 
assimilate him to its own being. Perhaps, in some way that 
eludes our schematic understanding, it will operate upon us 
in both respects at once; for every person's soul is divided, 
and to some degree everyone is groping toward the light and 
sinking into earthiness at the same time. Salvation and 
damnation are not clearly distinguished states, the one allotted 
to certain fortunate individuals and the other to certain un- 
fortunate ones. The two destinies represent the eternally war- 
ring factions of the human soul, with its simultaneous yearn- 
ing for the light and propensity for mud. To carry out Hera- 
clitus' metaphor we must think of mud as inflammable, but 
our thermonuclear discoveries have removed any difficulty 
on that score. In any case, the tenor of the metaphor of light 
and mud is ourselves, and the central question is, which of 
them we shall most essentially be. Our choice, which in some 
sense is made anew at each waking moment, develops a 
special corollary in the day of the great conflagration. For 
the issue then is, to each conscious being, whether he is to be 
flame or cinder. A literal restatement of the metaphor would 
be misleading ; each one who would find the meaning must do 
so by looking to his own depths and resources. 

At all events, when the ultimate cosmic situation is con- 

[ 81 ] 


sidered in terms of the perennial choice that Hes before each 
conscious individual, the appropriateness of Frs. 71, 72, and 
73 to the Heraclitean philosophy becomes more evident. We 
do not know the original context of Fr. 71, but we may assume 
that liars and false witnesses were mentioned merely as in- 
stances, and not the most important instances, of men choos- 
ing the downward way. The essence of the downward way, in 
what may be called (descriptively, not melodramatically) the 
catastrophic perspective, is to become a victim of the Event 
through attachment to the temporal things and attitudes that 
it will destroy. The fire as it advances is at once physical 
fact and divine judge. It judges and consumes whatever per- 
sons, and whatever aspects of each person, have failed to rise 
to its own condition of fiery activity and bright unrestricted 
awareness. To some extent, indeed, everyone and everything, 
by hardening into its own selfhood, has failed to become a 
pure dancing spurt of flame, and thus stands under the judg- 
ment. To become flame or to become cinder is the inescapable 
and constant dilemma to which every moving being must and 
does make his small contributory response at every moment.,^' 

82 ] 


80. Thinking is common to all. 

81. Men should speak with rational awareness and thereby \ 
hold on strongly to that which is shared in common — as \ 
a city holds on to its law, and even more strongly. For all ' 
human laws are nourished by the one divine law, which 
prevails as jar as it wishes, suffices for all things, and yet is 
something more than they. 

82. The people should fight for their law as for their city wall. 
8j. Law involves obeying the counsel of one. 

84. To me one man is worth ten thousand if he is first-rate. 

85. The best of men choose one thing in preference to all else, 
immortal glory in preference to mortal goods; whereas the 
masses simply glut themselves like cattle. 

86. Gods and men honor those slain in battle. 

8y. Even he who is most in repute knows only what is reputed 
and holds fast to it. 

88. To extinguish hybris is more needful than to extinguish a 

8p. It is weariness to keep toiling at the same things so that 
one becomes ruled by them. 

po. Dogs bark at a person whom they do not know. 

pi. What mental grasp, what sense have they? They believe 
the tales of the poets and follow the crowd as their teachers, 
ignoring the adage that the many are bad, the good are few. 

p2. Men are deceived in their knowledge of things that are 
manifest — even as Homier was, although he was the wisest 
of all Greeks. For he was even deceived by boys killing 
lice when they said to him: ''What we have seen and 
grasped, these we leave behind; whereas what we have not 
seen and grasped, these we carry away.'' 

pj. Homer should be turned out of the lists and flogged, and 
Archilochus too. 

[ 83 ] 


P4. Hesiod distinguishes between good days and evil days, 
not knowing that every day is like every other. 

p^. The Ephesians had better go hang themselves, every man 
of them, and leave their city to be governed by youngsters, 
for they have banished Hermadorus, the finest man among 
them, declaring: ''Let us not have anyone amongst us who 
excels the rest; if there should be such a one, let him go 
and live elsewhere." 

p6. May you have plenty of wealth, you men of Ephesus, in 
order that you may be punished for your evil ways! 

p/. After birth men wish to live and accept their dooms; then 
they leave behind them children to become dooms in their 

Unlike Pythagoras, Heraclitus was not a founder of com- 
munities. His attitude toward community is forcefully ambiv- 
alent. In a profound sense the community is divine, and all 
human laws are nourished by the universal and all-sufficient 
divine law (Fr. 8i), which is the intelligence that ''steers all 
things through all things" (Fr. 120). On the other hand, any 
actual community is found to be made up largely of those 
moist and drunken souls that Heraclitus despises. To charac- 
terize the political aspect of Heraclitus' philosophy requires 
an equable recognition of these two opposing and mutually 
qualifying ideas. 

Frs. 95 and 96 give brief indication of an event that doubt- 
less had a sharp impact on the philosopher and confirmed him 
in his misanthropic cast of mind. Nothing is known about 
Hermadorus, nor about the grounds on which the Ephesians 
banished him; but there is no doubt concerning Heraclitus' 
judgment of the matter. And whatever personal chagrin and 
deprivation he may have felt at the banishment of a friend, it 
is evident that his thought must have been concerned with the 
more general political danger of which the incident was a 
reminder — the ever-present threat of collective mediocrity 

[ 84 ] 


against the outstanding individual. Dogs, after all, bark at 
anyone whom they do not recognize (Fr. 90). 

Mediocrity is given a more positive meaning in Fr. 88 
through the concept of hyhris. In the Introduction I have 
translated the word as ''flagrant self-assertion," and the 
phrase can be retained, although some further connotations 
may be noted. The more usual translation by such words as 
"pride" and "arrogance," although not wrong, is perhaps 
more liable to misunderstanding. Heraclitus himself seems 
to have been a proud and even an arrogant man, if we may 
credit the persistent ancient stories about him. He is surely 
not saying that his own kind of pride should be extinguished. 
The passion of distance was especially strong in him, and 
the quality of soul that is spoken of in Fr. 88 has to be under- 
stood as something opposed and antagonistic to the proud 
aristocratic struggle toward the light. Accordingly both 
Burnet and Fairbanks have seen fit to translate hyhris as 
"wantonness," and Fairbanks' version of the Fragment 
reads : "Wantonness must be quenched more than a confla- 
gration." This, at any rate, indicates an indispensable part of 
the idea. What the word connotes is spiritual slackness and 
arrogance together : it is the slackness that takes satisfaction 
in being slack. In every soul there is a tendency to become 
spiritually loose, riotous, and in this sense egotistical, and the 
tendency must be mastered if there is to be integrity and 
upward growth in either the individual or the community. 
The prized condition of soul and society alike is a tautness 
like that of a cord of the bow drawn back, or like that of the 
strings of a properly tuned lyre (Fr. 117). A loosening of 
the bow's cord prevents the archer from hitting the bull's- 
eye ; a loosening of the lyre's strings prevents the lyrist from 
producing musical sounds. Hyhris is the loosening of one's 
inner cord, with the result that self-control is lost and the soul 
becomes moist and slovenly. Incidentally, there is a typical 
paradox in the fact that Heraclitus employs the simile of a 
conflagration when the condition of which he is speaking is, 

[ 85 ] 


according to his usual metaphor, not a fiery but a watery one. 

Still, in acknowledging the foregoing distinction it must be 
recalled that for Heraclitus there is always another and con- 
trary side to every situation. Hybris, like everything else, is 
ambivalent, and can be seen in double perspective. While on 
the one hand it represents the ''moist" and arrogant vanity 
that characterizes inferior souls, it has a more universal signi- 
ficance too. For there is a sense in which everything tends to 
persist in its own specific being, and in which every person 
strives to retain and assert his own selfhood. Self-assertion, 
even flagrant self-assertion, is a universal characteristic ; it is 
what makes possible and inevitable the strife that gives a 
meaning to existence, and without which all things would 
cease to be (Fr. 27). On the human level it becomes more 
specifically egoistic ; there, in dramatic perspective, it supplies 
the tragic flaw (a/xaprta) which Aristotle regarded as the 
efficient cause of authentic tragedy.^ The central tragic in- 
sight, generalized beyond the boundaries of the stage, regards 
human self-assertion, which is to say human existence, as 
always in the long run self-terminating. Heraclitus would 
have agreed, but his agreement does not involve nihilism or 
melodrama or self-pity. For termination does not mean de- 
feat, except for those who are its unwilling victims. The wise 
and serene soul, moving with confidence and grace toward 
the light, accepts the temporality and the conditions of it by 
his own choice (cf. Fr. 45) ; and thus he makes himself not 
a defeated victim of change but a participant in the divine 
process of self-overcoming. 

Those who express and choose the better way, instead of 
slipping into the worse (granted that both ways are, in 
ultimate reckoning, necessary complementaries) are referred 
to in Frs. 84, 85, and 86. To die in battle — i.e., in a condition 
of intense activity — is far more honorable and excellent than 
to die in sluggish illness.^ The first-rate man (aptcrro?) is a 
dry, active soul ; he does not follow the crowd and their popu- 
lar slogans (Fr. 91), and in a good community he would be 

[ 86 ] 


the natural ruler (Fr. 84). But from the turgidity and con- 
fusion of actual communities he must lead his essential life 
apart, even as wisdom "stands apart from all else" (Fr. 7). 

Yet, while there is a sense in which he must stand apart, 
there is also a sense in which the superior man must never 
fall into isolation : he must maintain, through quickened per- 
ceptions and intelligence, an active and ever renewed kinship 
with the fiery Logos that encompasses him. Such kinship 
cannot be realized entirely in isolation; it is necessary to be 
''guided by what is common" (Frs. 2, 15), while at the same 
time correcting and challenging men's false apprehensions 
of it. The importance of this principle of shared yet directed 
experience finds political embodiment in Fr. 81. The Frag- 
ment begins with a pun upon the Greek expressions for ''in 
common" and "with rational awareness" f it then proceeds 
to introduce a word that has not appeared in any of the 
previous Fragments, but which is now employed in Frs. 81, 
82, and 83 — the word i^djoto?, law. Both Reinhardt and Gigon 
think that there is a distinction to be made between "all 
human laws" and "the laws of the city" — the former repre- 
senting the customs, precepts, and traditions, in short the un- 
written laws by which human life is largely carried on, while 
the latter are the codified laws. The distinction may seem to 
have some justification in the fact that the first translation of 
this word given by Liddell, Scott, and Jones is "that which 
is in habitual practice, use or possession." But while the word 
is there further said to mean "custom, usage," the more 
definite meaning is also given, "statute, ordinance made by 
authority." In any case. Kirk wisely rejects the distinction 
as misleading;* for the Greeks regarded the codified laws as 
growing out of and supported by the unwritten laws, and both 
of them are said by Heraclitus to participate, although im- 
perfectly, in the "one divine law." 

Although the one divine Nomos is not essentially different 
from the one divine Logos, there is a suggestive difference in 
connotation, determined by the difference of the metaphors 

[ 87 ] 


that are represented. The doctrine that is brought in with the 
new word contains an indispensable part of the full truth — 
namely, that a community can thrive only if there is some 
mutuality of rational awareness among the citizens, and that 
such rational and communal awareness can come only by 
drawing upon the divine source. 

The comparison in Fr. 82 between the law and the city 
wall is not accidental. The wall of a city in ancient times was 
far more than bricks and mortar; it was a kind of magical 
encirclement, representing and guaranteeing some kind of 
supernatural protection. Much later, in Vergil's Aeneid, there 
are two oblique but suggestive references to the importance 
of the city wall. The better known of them involves the 
episode of the wooden horse, filled with Greek warriors, which 
by treachery gained entrance into the walls of Troy and led 
to the city's downfall. It seems unlikely that in actual fact the 
Trojans would have been deceived by so gross and strange a 
maneuver, and the real question comes to be how such a 
fantastic story might have arisen. The legend could well 
have grown out of the exaggerated emphasis laid upon the 
protective power of the city wall, with the resulting sense that 
the safety and autonomy of the city are lost when this protec- 
tive encirclement has been violated. The other Vergilian 
reference occurs in the account of how Queen Dido won by a 
stratagem her right to build a kingdom in North Africa. A 
local potentate named Byrsa jokingly granted her as much 
land as could be encompassed by a bull's hide, and Dido met 
the condition by having her men cut the hide into very thin 
strips, whereby a considerable area could be encompassed, on 
which to build the city of Carthage. It is impossible to trace 
the origin of the playful legend ; and the fact that hyrsa is the 
Greek word for bull's hide proves nothing, because the name 
of an anonymous North African could have been invented to 
suit the story.^ But the bull was a sacred animal to worshipers 
of Dionysus and of Mithra, perhaps also to some other sects, 
and it was likely enough that the bull might have become 

[ 88 ] 


associated with the divine power guarding the boundaries of 
a city. Nothing is proved by the two VergiHan episodes, ex- 
cept that they seem to add to the evidence of how defirly the 
ancient peoples regarded their city wall. What Heraclitus is 
doing in Frs. 82 and 81 is to connect the idea of the city 
wall with that of the city's law, and to connect the city's law 
(including what is unwritten) with ''sharing in common" 
and, by the help of the pun, with "rational awareness." 

The remaining Fragments of the group do not require 
special comment. Fr. 97 brings the group to a close on a note 
of somberness — not a gratuitous pessimism, but a naturally 
resultant expression of Heraclitus' general philosophy. The 
first clause states something rather close to the Freudian 
doctrine of the death-wish : that even at birth we accept, in 
an unconscious layer of the mind, the fact of coming death. 
Still we push on, in spite of the dark recognition, and we think 
we strive toward the light although darkness is no less a real 
part of our unconscious attachment. Moreover, along with the 
open will to survive and the hidden attachment to death, there 
is also the urgency to reproduce, and the expression of it 
leads to a renewal of the problem and the paradox in our 
offspring. A frank, clear awareness of mankind's common 
destiny and doom is what marks alike the philosophy of 
Heraclitus and the crucial moments in the greatest Greek 

[ 89 ] 


p8. Opposition brings concord. Out of discord comes the 

fairest harmony, 
pp. It is by disease that health is pleasant; by evil that good is 

pleasant; by hunger, satiety; by weariness, rest. 
100. Men would not have known the name of justice if these 

things had not occurred, 
loi. Sea water is at once very pure and very foul: it is drink- 
able and healthful for fishes, but undrinkable and deadly 

for men. 
102. Donkeys would prefer straw to gold. 
10^. Pigs wash in mud, and domestic fowls in dust or ashes. 
104. The handsomest ape is ugly compared with humankind; 

the wisest man appears as an ape when compared with a 

god — in wisdom, in beauty, and in all other ways. 
10^. A man is regarded as childish by a spirit, just as a boy 

is by a man. 
106. To God all things are beautiful, good, and right; men, on 

the other hand, deem some things right and others wrong, 
loy. Doctors cut, burn, and torture the sick, and then demand 

of them an undeserved fee for such services. 
108. The way up and the way down are one and the same. 
I op. In the circle the beginning and the end are common. 
no. Into the same rivers we step and we do not step. 

111. For wool-carders the straight way and the winding way 
are one and the same. 

112. The bones connected by joints are at once a unitary whole 
and not a unitary whole. To be in agreement is to differ; 
the concordant is the discordant. From out of all the many 
particulars comes oneness, and out of oneness come all the 
many particulars. 

11^. It is one and the same thing to be living or dead, awake 
or asleep, young or old. The former aspect in each case 

[ 90 ] 


becomes the latter, and the latter again the former, by sud- 
den unexpected reversal. 

114. Hesiod, whom so many accept as their wise teacher, did 
not even understand the nature of day and night; for they 
are one. 

11^. The name of the bow is life, but its work is death. 

An intellectual alertness, which is so essential a charac- 
teristic, to HeracHtus, both of the cosmic fiery Hght and of the 
individual intelligence that strives to become like it, is also 
demanded of anyone who would read Heraclitus' chiseled 
remarks with understanding. A reader who falls into static 
interpretations and stereotyped associations will miss a part 
of the meaning of almost every utterance. Man learns to 
exercise his mind by shaping his experiences — perceived, re- 
membered, and imagined — into concepts, and this is a useful 
and clarifying procedure so far as it goes ; but then as a kind 
of mental safety-play he tends to regard his network of con- 
cepts as equivalent to truth itself, and thus he disposes himself 
to shut out of serious consideration any intuitive possibilities 
that do not have a fixable relation to the conceptual system 
(cf. Fr. 63). Heraclitus challenges our customary concepts 
and our customary methods of forming and relating con- 
cepts by a sometimes startling use of what Asclepius calls his 
"symbolical and gymnastical" style of thinking and writing.^ 
The most characteristic difficulty in Heraclitus' philosophy 
lies in the demand which it makes upon its hearers to trans- 
cend the ''either-or" type of thinking and to recognize in each 
phase of experience that a relationship of "both-and" may be 
present in subtle ways that escape a dulled intelligence. Hera- 
clitus' thought moves not by exclusion but more characteris- 
tically by coalescence, and always with a sense of otherness. 
To him nothing is exclusively this or that ; in various ways 
he affirms something to be both of two disparates or two 
contraries, leaving the reader to contemplate the paradox, the 
full semantic possibilities of which can never be exhausted 

[ 91 ] 


by plain prose statements. The upward and downward ways 
are contrary and yet one ; the human soul is destined and yet 
is faced with the ever-present choice between up and down; 
the soul originates out of fire (Fr. 28) but it originates out of 
water (Fr. 44) ; time is eternal and yet time, like everything 
else, must come to a death which is also a rebirth ; God is at 
once universal process, the intelligence that steers the process, 
the model by which a wise man will guide himself (Fr. 106) , 
and a child idly moving counters in a game. To be sure, the 
logicizing intellect will undertake to analyze each of these 
paradoxes into its elements, explaining in just what pair of 
respects, or in what pair of circumstances, or from what 
opposite points of view, something is at once such and not- 
such. But Heraclitus regards the paradox itself, and not its ) 
logical transformation, as more truly representing the real ) 
state of affairs. 

It is this acceptance of the ontological status of paradox — 
an acceptance, that is to say, of the view that paradox lies 
inextricably at the very heart of reality — that gave Heraclitus 
his ancient reputation for obscurity; and it is what stamps 
his philosophy with a different quality from that of an Ionian 
scientist such as Anaximander, or even from that of an 
Eleatic or a Pythagorean. To be sure, gnomic utterances were 
nothing new. The sense of the cryptic was strongly marked 
in the rising poetic consciousness of the sixth century B.C. 
Philosophers, however, were expected to be able to catch a 
glimpse of something steadier and more permanent. It was 
gradually coming to be recognized that their task was to 
look for the apxq (the fundamental principle of things, which 
is also their beginning) and the X0709 (the fundamental prin- 
ciple, which is also the meaning that can be uttered or that 
comes to the receptively wise man as if divinely uttered to 
him), and that these would somehow explain the manifold 
and fluctuating particulars. The earlier scientific philosophers 
of Miletus, Thales, and his followers, had sought a principle 
amid the elements of the physical world; although the most 

[ 92 ] 


Speculative of them, Anaximander, pushed his imagination 
farther and attained to a transcendental notion — the idea of 
an infinite ''Boundless," from which all physical manifesta- 
tions come into existence and back into which they vanish. 
Pythagoreans and Eleatics, with different orientations, ex- 
plored the farther reaches of transcendental method — an 
exploration that Plato was later to pursue with an unexampled 
many-sidedness of imaginative approach. The peculiar dif- 
ference that marks the Fragments of Heraclitus grows out 
of his refusal to accept either of these two recognized types 
of explanation. Although his philosophy shows a certain re- 
semblance to both the naturalistic and the transcendental 
philosophy in particular respects, he is not content to take 
either a perceptible physical substance or a postulated meta- 
physical entity as the be-all of existence. As remarked in 
Chapter iii, the fire of which he speaks is neither strictly 
physical nor strictly metaphysical; it is physical and meta- 
physical together ; for it is the feeding flame, perceptible both 
to outward sight and as inward exuberance, and at the same 
time it is the universal fact of perpetual change. The concrete 
and the absolute meanings of it overlap, and consequently no 
such clear-cut doctrine as the view of Thales that everything 
is water, or the atomic theory that was to come with Leucip- 
pus and Democritus, or an Eleatic definition of real being as 
timeless, can ever emerge in him. Heraclitus is unwilling to 
pursue clarity at the cost of distorting the truth of things 
as he finds it in the confused, shifting, and paradoxical mani- 
festations throughout experience. It is the bold attempt to 
put this elusive character of truth into some kind of nearly 
adequate language that gives to the Fragments their charac- 
teristic temper and characteristic difference which caused 
their author to be called ''the dark one." 

As an introduction to the more pointedly paradoxical state- 
ments that follow (Frs. io8ff.) a glance may be given to Frs. 
98-107, which record somewhat more commonplace observa- 
tions about the relativity of perceptual and valuational judg- 

[ 93 ] 


merits. Such statements of experiential relativity do contain 
seeds of paradox to be sure, but actually they do no more 
than set certain conditions upon which possibilities of oppos- 
ing points of view may rest. The development of such opposi- 
tion into the succinct semantic tension that is paradox comes 
about when a way is found of declaring the opposing terms 
to be not only mutually related, but somehow identical. It can 
easily be observed how this has been done in a variety of ways 
in Frs. 108-115, whereas the earlier Fragments offer more 
explicit comparison of various pairs of opposites, stating 
either their causal or epistemological involvement with each 
other or their axiological relativity. 

In the case of a writer like Heraclitus, who employs figures 
of speech not for prettification but as a means of exploring 
and adumbrating some of the more hidden aspects of reality, 
it is helpful to examine the ways in which those figures of 
speech function semantically ; and one of the most indicative 
questions of this sort has to do with the way in which paradox 
and metaphor are related. For each of these two tropes tends 
to involve the other whenever it is employed seriously and 
not superficially. By ''superficial" and "serious" I mean what 
I have elsewhere designated by the words "surface" and 
"depth. "^ There is both a paradox of surface and a paradox 
of depth, and there is both a metaphor of surface and a meta- 
phor of depth. A surface paradox is correctly defined as a 
"seeming contradiction" ; it can be explained away by making 
the proper logical qualifications. A surface metaphor is a 
tabloid simile; it declares in effect that something is like 
something else, but it does this more briefly by dropping out 
the word "like" and saying that the something is the some- 
thing else. Depth paradox and depth metaphor, on the other 
hand, always tend to occur somehow in mutual association. In 
the most characteristic utterances of Heraclitus, in particular, 
paradox and metaphor show something of this mutuality; 
with the result that his paradoxes and his metaphors are 
essential, not superficial — you cannot unsay them. You can 

[ 94 ] 


unsay a paradox such as "Christ was not a Christian" or 
'Trend was not a Freudian" : you can explain it away by 
remarking that Christ or Freud would not have identified 
himself with the views or practices of his self-professed fol- 
lowers. Again, you can unsay a metaphor such as ''He is a 
fox" ; for this is but an abbreviation of the simile, "He is as 
crafty as a fox." When, on the other hand, Heraclitus speaks 
about the cosmic fire, it is impossible to drop the figurative 
language without essential loss ; the relation between semantic 
vehicle and semantic tenor is organic, not mechanical. And 
when he declares a basic paradox, such as the identity of the 
upward and downward ways, it is not possible to explain this 
entirely away by a reshuffling and redefinition of logical com- 

In short, if metaphor and paradox are to serve a meta- 
physical purpose, each must to some degree involve the other. 
If metaphor is employed without a touch of paradox, it loses 
its radically metaphoric character and turns out to be virtually 
no more than a tabloid simile. If paradox is employed with- 
out metaphor, it is no more than a witticism or sophism. The 
double principle is so important for understanding Heraclitus* 
more obscure Fragments that it is worth while to look at each 
of the two complementary aspects separately. 

(I) First, then, metaphor, if it is semantic rather than 
merely grammatical, involves paradox. A merely grammatical 
metaphor is one that could be restated as a simile without 
essential loss — that is, without semantic alteration, without 
distortion of meaning. To call a man a pig is to say in effect, 
"He eats like a pig" ; and in changing the sentence from "He 
is a pig" to "He eats like a pig" there is no appreciable change 
of meaning. The simile sets in explicit relation the two ele- 
ments of discourse, man and pig, which even in the metaphoric 
version the speaker must really have kept distinct and uncon- 
fused in his mind. He could do this because he knew, inde- 
pendently of the specific attempt to connect them, what a man 
is and what a pig is. Hence the metaphor in such a case 

[ 95 ] 


differs from the corresponding simile in hardly more than a 
verbal way ; the difference is but one of grammatical formula- 

Consider, by contrast, any of the metaphors by which 
Heraclitus or any other philosopher attempts to characterize 
the fundamental reality : that it is an ever-living fire (Fr. 29) , 
that it is wisdom (Fr. 119) or intelligence (Fr. 120) or 
reason (Frs. i, 2, 118), or (passing quite beyond the Hera- 
clitean context) that it is love. A metaphoric assertion of such 
a kind cannot be reduced to a simile without a subtle change 
creeping into the meaning. Suppose that instead of saying, 
''God is reason," one were to say, "God is like reason." Al- 
though there is an apparent gain in logical precision here, the 
appearance of clarity is misleading, for a certain unspoken 
assumption has been unwittingly introduced. In uttering the 
simile we assume that we know independently what God is 
and what reason is, and that it is possible to discover a rela- 
tion of similarity between the two ideas independently con- 
ceived. In our previous example, of calling a man a pig, the 
condition is adequately met. We know by independent sets of 
experiences what a man is and what a pig is ; we are then able 
to say that in a particular instance we can discover in what 
is undeniably a man certain specific similarities to what can 
be independently recognized as a pig. The situation is mainly 
one of comparison, and that is why it is possible to restate the 
apparent metaphor, "He is a pig," in the form of a simile, 
"He eats like a pig." Here, obviously, it is the simile, not 
the metaphor, that expresses more exactly what is meant. 

But in the three metaphysical metaphors just cited no 
parallel situation is present. The subject-term — whether we 
use the word "It" or "God" or "Reality" or whatever else — 
is not known independently of the three predicates that are 
successively attached to it. One does not first know the 
Ultimate and then find that it bears a certain similarity to 
intelligence, to fire, to love, etc. Such imperfect knowledge 
as we can have of the Ultimate is had through such attributes, 

[ 96 ] 


by attending to the overreach of meaning that each of them 
can suggest, beyond the trade-meanings of ordinary expe- 
rience. As a perceptive man stumbles along on his pilgrimage 
through experience he discovers that some qualities and some 
relationships are semantically more promising than others, in 
that they tend to suggest fuller meanings than lie on the 
surface or than are grasped on early acquaintance. The 
serious percipient finds himself led on, then (partly but not 
exclusively influenced by conditioning factors from outside), 
to regard certain aspects of his world as divine ; and he may 
attempt to verbalize his insights by such metaphors as "God 
is the bright sky" or ''God is fire," or at a more developed 
stage, ''God is intelHgence" or "God is love." But each of 
these metaphors, whether it grows out of primitive or more 
matured experience, is a paradox. To identify the ultimate 
Reality with some finite part of itself is to say, in effect, "The 
Infinite is the finite." Yet in speaking of Divinity a use of 
radical metaphor is unavoidable. Such metaphor must always 
be radical, not merely grammatical ; which is to say, it must 
work mainly not by comparison, like a simile, but by insight 
and transcendental probing. To say that fire or wisdom or 
love is divine is to start with one of these three specific and 
familiar phenomena and to become reverently open, attentive, 
and responsive to the values and suggestions of transcendental 
association that may, without looseness or irresponsibility, be 
found in it. In pursuing this eductive course — this semantic 
passage from the more everyday and familiar to the darkly 
hinted at — we employ a metaphor which by literal standards 
says too much. We declare, "Fire is divine," or we turn the 
sentence around and declare, "God is fire." And the result, in 
either direction — if the implications of the transcendental term 
are not ignored or caricatured — is flagrantly paradoxical. 

(II) But if serious metaphor thus involves paradox, it is 
also conversely true that serious paradox involves metaphor. 
This converse involvement can be shown by examining 
paradox in the same double manner in which we have ex- 

[ 97 ] 


amined metaphor — looking first at ordinary, superficial 
paradox and then observing how radical paradox differs 
from it. Now ordinary paradox, or paradox of surface, as has 
been said, is merely a trick of speech whereby a point can be 
made more wittily and effectively. "Christ was no Christian," 
"Freud was no Freudian," "Nothing is more fatiguing than 
leisure," or Chesterton's "Nothing is so miraculous as the 
commonplace," etc. — such paradoxes as these may grace an 
evening's conversation and give the perpetrator a reputation 
for cleverness, but actually they depend for their effect upon 
a more or less deliberate confusion between two connotations 
of a word — e.g., between Christianity as it is practiced and 
Christianity defined as adhering to the principles of its 
founder, or between genuine leisure and the sheer inoccupancy 
in which one burdens oneself with trifles, or between the 
dramatic sort of miracle that involves a breaking away from 
the commonplace and the ultimate metaphysical miracle that 
existence and order should be instead of nonexistence or 

By contrast, a radical and serious paradox does not hang 
upon a removable confusion, but is demanded by the com- 
plexity and inherent ambiguity of what is being expressed. 
Fr. 1 08, one of the most familiar sayings of Heraclitus, offers 
an example. To say that the way up and the way down are 
one and the same is, if taken literally, false and absurd ; such 
a statement is not even a paradox, but merely a piece of non- 
sense. Yet it is evident that Heraclitus was attempting to 
declare something deeply significant here. How is it possible 
to understand the identification as making sense rather than 
nonsense? Obviously, as every reader immediately perceives, 
the statement must be taken metaphorically. Paradox implies 
metaphor, in that the upward way can be identified with the 
downward way only with respect to certain metaphoric con- 
notations of each. Now in Heraclitean context the upward 
way is the way from earth to water to fire, and the downward 
way is the reverse. But these two directions of transformation 

\ [ 98 ] 


occur not only in physical nature but also in the life of the 
self. Moreover, the phrase *'one and the same" must also be 
interpreted with some metaphoric flexibility : it evidently 
means to say that the two contrary processes are both going 
on all the time, and that their continual and varying tension is 
what makes existence and life possible. Thus a meaningful 
interpretation of the paradox in Fr. io8 serves to reveal 
several metaphors that are latent in it. And in general, with 
differences in detail, much the same thing will be found to 
be true in all of Heraclitus' paradoxes. 

The question of the use of figurative language comes up 
in a rather special respect in Fr. iii, 'Tor wool-carders the 
straight way and the winding way are the same." Naturally, 
the literal meaning of a passage should be established first, in 
order to avoid ill-grounded inferences as to the symbolical 
tenor. Hippolytus, when quoting the present Fragment in 
The Refutation of All Heresies, explains the literal situation 
thus : "The circular movement of an instrument in the fuller's 
shop called 'the screw' is at once straight and curved, in that 
it revolves upwards and circularly at the same time."^ If this 
were the double motion that Heraclitus had had in mind, a 
simultaneity of a straight and a curved propulsion, as a pair 
of component movements, it would not be easy to see just 
what symbolic reference might have been intended. Hera- 
clitus, however, shows no evidence of possessing a developed 
geometrical imagination, and would probably not have 
thought in terms of component movements. Moreover, the 
instrument called "the screw," although it evidently was em- 
ployed by fullers at the time when Hippolytus wrote, prob- 
ably had not been invented seven centuries earlier in Heracli- 
tus' day. Then, too, it should be noted that the word o-koXl6<; 
means not only "curved" but also, and more loosely, "wind- 
ing, twisted, tangled." All in all, although it is not certain 
just what process of wool-carding was current in the earlier 
period, it seems probable that the meaning of the Fragment 
depends in part upon a pun in the word for "straight," which 

[ 99 ] 


In Greek carried a moral connotation, ''right" or ''straight- 
forward," as well as the geometrical one. On one level of its 
meaning, then, the Fragment could be taken to say, "For 
wool-carders the right way (i.e., for them as craftsmen) is the 
winding way." But then, just as one paradox emerges from 
taking the first adjective to mean "straight," so another 
paradox also emerges from the moral overtones of the second 
adjective, which can mean not only "winding" but also "un- 
righteous, unjust, deceptive." Heraclitus employs a word- 
play of this kind with serious intent, for to his mind the 
occasional duplicity of language has an intimate connection 
with, and reveals, the duplicity and paradoxicality of the thing 
referred to. 

Another Fragment in the group — Fr. 115, "The name of 
the bow is life, but its work is death" — also depends upon a 
pun to make its point. One of the two Greek words for "bow" 
is )Sto9 with the accent on the final syllable, whereas one of the 
two main words for life is )8to9 with the accent on the initial 
syllable. The linguistic accident, whereby a death-dealing in- 
strument has a name so similar to a word for life, seems to 
Heraclitus to be significantly related to the great paradoxical 
fact that life and death are but two intertwining aspects of the 
same thing, both of them being present and producing an 
ever-changing tension in every phenomenon. 

Finally there is the compact group of expressive ambigui- 
ties in Fr. 109, "In the circle, the beginning and the end are 
common." All four terms need to be examined here. The least 
important of them, although not to be overlooked, is the 
suggestion of a pun which the word fwo? always appears to 
carry for Heraclitus — an interplay of the two meanings, "in 
common" and "with understanding" (cf. Note to Fr. 81). 
The ideas of "the end which is also a beginning" and the "be- 
ginning which is also an end" are a pair of archetypal para- 
doxes that have had a wide appeal to thoughtful men. One set 
of connotations is temporal, suggesting two ways, a forward 
and a backward, of looking at the passage of time. In spatial 

[ 100 ] 


representation their precarious truth is manifested most 
clearly in the geometrical figure of the circle. Most important 
of all, there is the deeper symbolic sense in which beginning 
and end are, or can be and should be, related. A philosopher 
slightly later than Heraclitus, Alcmaeon of Crotona, expresses 
this further meaning in his remark, "Men perish because they 
cannot join the beginning with the end."* Alcmaeon's words 
for beginning and end are dpxn and reXos ; the former of which 
connotes also "first principle," the latter "governing aim." 
Heraclitus also says dpxrj, but his word Trepa?, signifying end 
in the sense of limit or boundary, does not in itself carry the 
second additional connotation. Nevertheless, it receives some- 
thing of the fuller symbolic meaning from the connotations of 
the other words in the sentence, as well as from the realistic 
awareness, so characteristic of the ancient Greeks, that the 
idea of limit is always especially pertinent to human affairs. 
The overtones of the word "circle" are widely shared in 
different ages and different cultures; for here a geometrical 
figure is immediately presented to the imagination, inde- 
pendently of any special linguistic usage. In many an ancient 
culture the idea of the circle could mean at once the spatial 
figure, a cycle of events in time, and that serene condition of 
affairs in which there is regularity of movement about a 
still central point, whereby the initiating principle of life and 
its guiding end are brought into harmonious union. 

[ loi ] 


^ ii6. The hidden harmony is better than the obvious. 
11/. People do not understand how that which is at variance 
with itself agrees with itself. There is a harmony in the 
bending back, as in the case of the bow and the lyre. 
ii8. Listening not to me but to the Logos, it is wise to 
acknowledge that all things are one. 

/^iip. Wisdom is one and unique; it is unwilling and yet will- 

\i^g to be called by the name of Zeus. 

C120. Wisdom is one — to know the intelligence by which all 

V things are steered through all things. 

121. God is day and night, winter and summer, war and 
peace, satiety and want. But he undergoes transforma- 
tions, just as *****^ when it is mixed with a fragrance, is 
named according to the particular savor [that is intro- 
duced'] . 

122. The sun will not overstep his measures; if he were to 
do so, the Erinyes, handmaids of justice, would seek him 

12^. All things come in their due seasons. 
124. Even sleepers are workers and collaborators in what goes 
on in the universe. 

So CLOSELY has Heraclitus' name been associated in Western 
philosophical tradition with the related themes of change and 
paradox, that there has often been a tendency to overlook the 
peculiar emphasis which he gives to the unity, in a qualified 
and paradoxical sense, of all things. Writers on ancient 
philosophy have sometimes been tempted to schematize their 
material by contrasting Parmenides and Heraclitus as repre- 
senting extreme opposite tendencies, thereby producing a vast 
philosophical conundrum which the so-called ''later natural- 
ists" — Empedocles, Anaxagoras, Leucippus and Democritus 

[ 102 ] 


— undertook to solve by their diversely ingenious types of 
metaphysical integration. Of course there is a large half- 
truth in this way of envisaging the matter : for it cannot be 
denied that Parmenides is above all else the philosopher of 
unity and permanence, who regards every apparent evidence 
of variety or change as ipso facto illusory ; nor can it be denied 
that Heraclitus accepts variety and change as basic and in- 
expungeable facts about the real nature of things. Unques- 
tionably the opposition between the philosophers is there for 
all to see. Nevertheless, the resulting stereotype, which repre- 
sents them as doctrinally contrasted and nothing more, can 
be grossly overdone. Relatively few qualifications need to be 
made for Parmenides, no doubt ; but the philosophy of Hera- 
clitus is too subtle, manifold, and shifting to be defined in 
such static terms. Granted that variety and change constitute 
a main theme, perhaps even the main theme for Heraclitus, 
there is nevertheless a second theme, running contrapuntally 
throughout the doctrine, which is equally indispensable — the 
theme expressed most plainly in Fr. 118, that "all things 
are one." It is misleading to call Heraclitus a pluralist with- 
out adding that he is somehow a monist as well, or to stress 
his doctrine of change, chance, and strife without adding 
that these characteristics, real and basic though they are, 
exist somehow counterbalanced by a tendency toward order, 
pattern, and harmony, which is equally inherent in what we 
must call (knowing that words fail us here) reality. 

Still, the mode of balancing requires careful attention and 
careful statement. The peculiar sense in which Heraclitus 
accepts cosmic oneness as a reality cannot be indicated by 
any of the usual philosophical labels. Even so learned a Her- 
aclitean scholar as Hermann Diels made the mistake, some fif- 
teen years after compiling his valuable collection of the Frag- 
ments, of putting forward an interpretation of Heraclitus as 
a metaphysical dualist. In his article on the philosopher in 
Hastings' Encyclopaedia of Religion and Ethics Diels es- 
pouses a distinction between what he calls the **husk" and the 

[ 103 ] 


"kernel" of Heraclitus' doctrine — the former term referring 
to the fiery world of change, the latter to the ultimate one- 
ness of the Logos. He even goes so far as to support his 
quasi-Platonic interpretation by some apparently biased trans- 
lation. For example, Fr, 120 ends with the phrase, iravra 
hta TrdvTcov, which characterizes the cosmic intelligence as 
steering ''all things through all things" — an idea of typical 
Heraclitean complexity; Diels, however, changes his trans- 
lation in the Encylopaedia article to read, ''which knows and 
governs all things." Now there is a good deal of connotative 
difference between this phrase and the one that Heraclitus 
uses, and the difference becomes even more significant when 
it is realized that the verb which I have translated "are 
steered" (to avoid an excessive suggestion of individual lib- 
erty) might equally well, or almost equally well, be trans- 
lated "steer themselves." For the difference between the 
Greek passive and the Greek middle voice is not grammati- 
cally indicated in certain tenses, and a number of Heraclitus' 
phrases suggest an inseparable coalescence of the two kinds 
of idea. The intelligence — which is also to say the thunder- 
bolt (Fr. 35) — that does the steering operates on particular 
things, including human individuals, from within as well 
as from without; and the result, as several earlier chapters 
have shown in different perspectives, is a universe in which 
unity, diversity, self-direction, chance, and ambiguity all 
commingle. But Diels' mistranslation enables him to reach 
the strange and unacceptable conclusion that "Heraclitus 
comprehends, as exactly as his opponent Parmenides, who 
only partly understood him, noumena and phenomena, truth 
and illusion, in his system."^ It is one thing to compare, 
another to equalize, such diverse thinkers as Heraclitus and 
Parmenides. The kinds of cosmic unity for which they re- 
spectively stand are radically different. 

Diels' dualistic interpretation of Heraclitus causes him to 
belittle such statements as Fr. 24 ("Time is a child moving 
counters in a game") and Fr. 40 ("The fairest universe is 

[ 104 ] 


but a heap of rubbish piled up at random") as referring only 
to "ephemeral experience," which he thinks represents to 
Heraclitus ''the mutable, inconsistent, unconscious, and 
childish world of change." But to Heraclitus' notion all ex- 
perience is ephemeral, and all things are mutable and partly 
or potentially inconsistent. Diels, however, connects Frs. 24 
and 40 with Heraclitus' denunciation of polymathy (Fr. 6), 
and he opposes the three Fragments as a group to the state- 
ments that "wisdom stands apart from all else" (Fr. 7) and 
that self-knowledge is needful as the way of wisdom (Frs. 8, 
9). Now it is true that Heraclitus is against polymathy, or 
mere extensive learning for its own sake, and that he regards 
the way of wisdom as lying through self-knowledge, or at 
least as inseparable from it. In Fragments 6 and 9 he is 
asking the important question, "What is the best way of 
thinking?" or, "In what does an intellectually superior atti- 
tude toward the universe consist?" And the answer which 
he affirms is quite different from that which Diels attributes 
to him. For while wisdom is "apart" from other things, in 
the sense that it is intrinsically more valuable than they, 
such wisdom is genuine only if it is a fiery activity, able 
to reflect the changing facets of the world itself — "for the 
moving world can only be known by what is in motion" 
(Fr. 43)- 

The unity of things as Heraclitus understands it is a subtle 
and hidden sort of unity, not at all such as could be ex- 
pressed by either a monistic or a dualistic philosophy. The 
oneness of things, or rather their mutual attunement, cannot 
exist or even be conceived apart from their manyness and 
discord. The wisdom that steers all things through all things 
(overtone: "the wisdom by which all things steer themselves 
through all things") is something that cannot be expressed 
without paradox. To call it "God" or "Zeus" is at once 
necessary and misleading (Fr. 119). It is necessary, because 
there is no way in which to refer to or think about the ulti- 
mate cosmological problem of the relatedness of all things 

[ 105 ] 


except by employing the least inadequate symbols that we 
possess. On the other hand, any divine name is misleading, 
because of the doctrinal and mythological associations that 
it almost inevitably conjures up in people's minds. No at- 
tempt to characterize the ultimate unity of things, or the 
power and tendency toward unity, can possibly succeed, and 
yet man's inquiring mind cannot permanently abandon the 

The transcendent neutrality of ultimate reality is indicated 
in Fr. 121, where unfortunately a crucial word, here indi- 
cated by asterisks, is missing in the extant manuscript of 
Hippolytus — presumably dropped out accidentally by a copy- 
ist. Certain scholars, including Diels and Burnet, have arbi- 
trarily guessed that the missing word was ''fire," but there is 
no textual support for this, and I cannot see in it anything 
more than a sort of scholarly stock response: "Heraclitus, 
ergo fire." Moreover, there is no custom on record, so far 
as I am aware, either among the ancient Greeks or anywhere 
else, of naming fire according to the incense (or possibly, 
as some translators have it, the spices) that are thrown 
into it. A more plausible suggestion has been put forward 
by Hermann Frankel,^ who suggests that the missing word 
might have referred to the pure oily base with which dif- 
ferent perfumes were blended in making ointments, which 
would then be called by one name or another according to 
the resultant savor. The oily base (or possibly beeswax?) 
that was employed in such manufacture would of course have 
to be entirely pure, in the sense of not possessing any odor 
in itself. Consequently the point of the comparison would 
be (if Frankel's theory is right) that the ultimate reality 
(what can be called, and yet cannot be called, God or Zeus) 
is itself so pure and unparticularized that it does not possess 
any qualities whatever, thus being susceptible to any and 
all manifestations and changes. 

But can nothing more definite about the cosmic World-All 
be said? To regard it as having no characteristics, being no 

[ 106 ] 


sort of this as distinguished from that, is to reduce it to a 
virtual semantic zero. And yet Heraclitus is quite evidently 
endeavoring to say something, and he evidently attaches 
meaning to what he wants to say. Is there any way, then, 
in which ultimate reality can be characterized? What — to 
put the question as provokingly as possible — can be uttered 
about the Unutterable? One can only reply that while there 
are indications, the indications are scattered and sometimes 
in tensive opposition to one another, and that it would be 
a mistake to pursue any of them too far or in exclusion. 
There are indications of a kind of order, balance, and self- 
steering in the universe (Frs. 29, 35, 120, 122), which, al- 
though it never halts the changing passage of time, does give 
some kind of boundaries and character to universal change, 
so that to the discerning mind the change is never simply 
chaos. There are indications of a wisdom, intelligence, and 
reason (Logos) — not to be measured and judged by the 
petty and partisan standards of our human thought, but still 
profoundly real and effective (Frs. i, 2, 118, 119, 120). 
And in the first two Fragments of the present chapter there 
are indications of a cosmic order under the aspect of har- 

Although it may possibly be true, as Kirk declares, that 
the Greek word apiiovia did not acquire a musical meaning 
until the latter part of the fifth century, the point seems to 
be impossible to prove, and in any case Heraclitus would 
still be introducing a musical analogy in his reference to the 
lyre in Fr. 117. Already Pythagoras had taken the idea of 
musical harmony (connoting in Greek music probably not 
so much togetherness of sounds as a pleasing and fitting 
array of sounds) as the central symbol of his doctrine; and 
when discussing this doctrine with later Pythagoreans Soc- 
rates employs the word apfiovla.^ So all in all I think it 
is more justifiable to follow Fairbanks, Lattimore, and Kath- 
leen Freeman in translating dpfjLovta as "harmony" than 
to accept Burnet's word ''attunement" or Kirk's word *'con- 

[ 107 ] 


nexion." Heraclitus may perhaps be regarded as having 
taken the Pythagorean idea of harmony and added to it an 
observation of his own. Harmony, he adds, can exist only 
where there is contrast. There is no harmony of a single 
note, there is significant harmony only where there are ''op- 
posing tones" which are resolved.* Musical harmony involves 
the overcoming, but without the eliminating, of some musi- 
cal opposition. And in the larger sphere of human existence 
the same situation is found to occur. Harmonies and attune- 
ments between person and person, or between person and 
circumstance, are brought into existence out of diversity and 
potential strife. 

Moreover, there is not merely the contrariety of different 
musical notes, there is also a contrariety of tensions in the 
way in which the musical sound is produced. The strings 
of the lyre must be bent back and released, in order for the 
sound to come forth. Somewhat analogous, as Fr. 117 sug- 
gests, is the case of the bow. Pfleiderer sees in the conjunc- 
tion of these two similes a reflection of the familiar Hera- 
clitean paradox of life and death, and accordingly of the 
upward and downward ways.^ But the Fragment shows no 
intention of making a comparison between the bow and the 
lyre; the comparison is rather between each of these instru- 
ments and the tenor of the argument. Looking at the double 
simile in this way we can see more fully what the tenor is. 
In the case of both the bow and the lyre a string has to be 
pulled back and strained in order to be released : that is evi- 
dently the simple fact on which the point of the Fragment 
rests. But what is the result, in the two cases, of releasing 
the string? In the one case there is a hitting of the mark, 
in the other a production of musical sound. The connection 
of these two ideas, when each of them is taken metaphori- 
cally, is a basic Pythagorean theme, and it is one which 
was congenial to Greek ethical thought generally. A missing 
of the mark (dfjiapTia) is, according to Aristotle, the central 
flaw that precipitates tragedy;® and a human life to which 

[ 108 ] 


the metaphor may justly be appHed is also one that has 
fallen into disharmony — with other persons, with situations, 
and above all with itself, which is to say with its vocation. 
Contrariwise, the man who can hear and whose soul can 
echo "the music of the spheres" (as Pythagoras expressed 
his highest cosmological conception) is best equipped and 
readiest to know and recognize the truth. The idea implicit 
in the connection of bow and lyre, then, is presumably Pytha- 
gorean, or at least bears a striking resemblance to one of 
the highest tenets of Pythagoreanism ; but the "bending 
back" represents Heraclitus' own distinctive emphasis. 

The closing Fragment, Fr. 124, suggests a kind of reso- 
lution to the strongly tensive philosophy. Although it is 
better to be awake and to strive toward the light, yet even 
those who sleep and fall into darkness are still "workers and 
collaborators in what goes on in the universe." The idea is 
a favorite one with Stoic philosophers, and in fact it is 
Marcus Aurelius who has preserved the present Fragment 
by quoting it in his Meditations.'^ It is another expression of 
the paradox of the double perspective — ethical and cosmo- 
logical — that is so central to Heraclitus and the Stoic alike. 
In ethical perspective the choice between the two ways lies 
before each of us, and to the alert mind it is an urgent choice ; 
but in metaphysical perspective the upward and downward 
ways are both in process all the time, and all things even- 
tually steer themselves numberless times in both directions; 
so that, in Heraclitus' inaccurate but succinct phrase, the 
two ways are "one and the same." It is better to be a man 
of calm wisdom than a fluttering fool, and better for one's 
intelligence to be dry and bright than to be a victim of moist 
emotions ; nevertheless the foolish and the dissolute have their 
roles to play in the ever shifting universe, like everything 
else. The ultimate order of things is perfectly comprehensive 
and perfectly impartial; the countless individual particulari- 
ties, each with its partisan aims and partisan point of view, 
become of infinitesimal significance in their relation to the 

[ 109 ] 


whole. Every theory about the ultimate order, every attempt 
to make a statement about it, is bound to express one of 
those partisan and partial points of view ; from which it fol- 
lows that no theory of or statement about the cosmic wis- 
dom, ''by which all things are steered through all things," 
can ever do justice to the complexity and subtlety of that 
cosmic order itself. Even the word ''wisdom" is a somewhat 
hapless metaphor as applied to it. Heraclitus would have been 
in agreement, could he have known it, with the opening re- 
mark of the Tao Teh Ching : "The tao that can be understood 
is not the real ^ao." The ultimate reality, of which Heracli- 
tus is trying to speak, cannot be adequately represented 
either by a rationalistic concept such as "being" or "truth," 
or by a theological concept such as "Zeus" or "God," or by 
a naturalistic concept such as "fire" or "atoms." No word or 
image or idea can do it justice; but one of the least inade- 
quate ways of symbolizing it, indicating as it does both its 
interrelating power and its elusiveness, is the phrase that 
Heraclitus employs in Fr. ii6 — the hidden harmony. 

[ iio ] 


In the following appendices the references to 
books and authors are usually designated by 
the surnames of their authors or editors. The 
one exception is Doxographi Graeci, edited by 
Hermann Diels, which is called ''Dox.'' to dis- 
tinguish it from Diels' Die Fragmente der Vor- 
sokratiker. The fourth edition of Die Fragmente 
(1920) is called "Diels" and the fifth edition 
(1934), as revised by Walter Kranz, is called 
"Diels-Kranz." Bibliographical details for all 
books cited in the notes will be found in Ap- 
pendix B. 

























































































































































































































































































































124, 125 









































































1. Plato Laws iv, 715 E; quoted by Diels-Kranz as fr. 6 
under ''Orpheus" : ap^-qv re Kai Tekevrrjv Kal ixeaa Ttov ovtcov arrdvTOiV 

2. Strabo the geographer (xiv. i. 25) speaks of Thales as 
the first who wrote on KjtvcnoXoyia. It should be remembered that 
the word is derived from (f>v€Lv, "to grow," and that the conno- 
tation of spontaneous growth tends accordingly to be present 
in all early Greek discussions of nature (</>vcrt?). 

3. Anaximander's word is to arretpov, variously translated "the 
boundless," "the non-limited" (K. Freeman), "the infinite" 
(Burnet), and "the indefinite," My phrase, "boundless reser- 
voir of potential qualities," indicates the fuller significance of 
the word as I understand it. According to Anaximander, as re- 
ported by Diogenes Laertius (11. i), the Boundless is a basic 
principle {apxq) and basic element {aroixelov). Diogenes adds 
the explanation, which he appears to attribute to Anaximander, 
that the parts undergo change whereas the whole {to ttSlv) 
does not. 

An objection has been raised against the phrase, "boundless 
reservoir of potential qualities," on the ground that a clear idea 
of the potential as distinguished from the actual was not formu- 
lated until a century and a half later by Aristotle, and that the 
application of such words to Anaximander's doctrine is ana- 
chronistic. The warning should be heeded : any use of later ter- 
minologies for an earlier doctrine is likely to be misleading. But 
the same danger confronts us everywhere, more or less, when 
we try to express relatively primitive and coalescent ideas by 
means of a more sophisticated language. At least we can say 
that the idea of potentiality lay potentially and implicitly in 
Anaximander's notion of a reservoir of unmanifest qualities. 

4. Under the heading "Anaximander" Diels-Kranz lists this 
as fr. B-i, whereas in Diels it is listed as fr. 9. "Reparation": 
TtW. "Justice" : SiKr}. The word "therein" corresponds to no 
particular word in the Greek, but indicates the connoted relation. 

5. Anaximenes, fr. 2, in Diels-Kranz's "B" list, p. 95. "Soul" 
represents the word i(/vxri and "breath" the word Trvevfia. In the 

[ 113 ] 


present context irvevfxa evidently carries more of a physical mean- 
ing than the other word. Six centuries later, when St. Paul used 
the word 7rve{;/xa to mean "spirit," the relation between the two 
words had become virtually reversed. 

6. The four quotations from Xenophanes represent his frs. 
15, 23, 24, and 25 in Diels-Kranz's list, pp. 132-135. 

7. The question of the relation between Heraclitus and Hip- 
pasus of Metapontum is of interest, both generally and because 
of the attacks which Heraclitus makes against Pythagoras in Frs. 
128 and 136 (Appendix C) as well as in the accusation (Diog- 
enes Laertius viii. 6, rejected by Diels-Kranz as a quotation) 
that his philosophy is a mere accumulation of secondhand learn- 
ing. Three questions need to be considered. 

A. As to the partial similarity between the doctrines of the 
two philosophers. Aristotle declares : "Fire is the material prin- 
ciple according to Hippasus and HeracHtus" (Metaphysics i: 
984 3l, y; ci. De Caelo iii : 303 b, 12) . Diogenes Laertius : "Hip- 
pasus of Metapontum was another Pythagorean, who held . . . 
that the All is limited and ever in motion" (R. D. Hicks' trans- 
lation in the Loeb Classical Library edition of Diogenes Laer- 
tius: II, pp. 397-398)- 

B. As to the relative dates. Several scholars take Hippasus 
to be later than Heraclitus. Theodor Gomperz speaks of him as 
one "who followed in Heraclitus' footsteps" and calls him "an 
eclectic philosopher . . . who sought to reconcile the teaching of 
Heraclitus with that of Pythagoras" {Greek Thinkers, i, pp. 
146, 371). Zeller (p. 195) calls him a later Pythagorean who 
was influenced by Heraclitus. R. D. Hicks {loc. cit.) even places 
him as late as the fourth century B.C. On the other hand, Proclus 
describes him as "an early Pythagorean" {Commentary on Eu- 
clid, Friedlander ed., p. 426, cited in Kirk and Raven, p. 231). 
And probably Aristotle regarded him as Heraclitus' predecessor, 
for when mentioning the two men together he places the name 
of Hippasus first although that of Heraclitus was the better 
known. Finally, Suidas (Diels-Kranz, p. 143) declares: "Some 
have said that he [Heraclitus] was a pupil of Xenophanes and 
of Hippasus the Pythagorean." 

C. As to the charge that Hippasus was not a good Pythag- 
orean. The charge is made by lamblichus in his Lije of Pythag- 
oras. But since lamblichus was a devout Pythagorean, while 
Hippasus was something of an intellectual rebel (cf. Burnet, 
p. 94, n. 2) who moreover, according to legend, had been ship- 

[ 114 ] 


wrecked for having revealed the Pythagorean mystery of the 
incommensurability between the hypotenuse and side of a right- 
angled isosceles triangle, it would seem that the charge against 
him must be taken with caution. There remain the testimonies 
that he had at least been a member of the sect. Probably he 
was not a very orthodox Pythagorean, but he may possibly have 
provided some sort of loose link between certain teachings of 
the Pythagorean school and Heraclitus. 

8. The two lines of Parmenides' verses to which reference 
is made are the last two in fr. 6 in Diels-Kranz's *'B" list under 
"Parmenides" (p. 233). "To be and not to be" is expressed by 
the phrase, to iriXeiv re Kal ovK etmt. The final phrase, which might 
be referring to Fr. 117 of Heraclitus, is: navTiov 8e iraXLvrpoiro^ 

IcTTL K€\€vOo<S. 

9. Karl Reinhardt, Parmenides (1916), esp. pp. I55ff. 

10. The essay is to be found at the beginning of Book ix of 
his Lives and Opinions of Eminent Philosophers, in Vol. 11 of 
the Loeb Classical Library edition of Diogenes. Although the 
gossipy anecdotes that Diogenes likes to tell may have little if 
any relation to historical truth, there is no reason to doubt his 
information as to the place, time, and lineage of the philosopher ; 
nor is there reason to doubt the accuracy of his few direct quo- 
tations, most of which are represented in the list of accepted 

11. See Nietzsche, Gesammelte Werke (Musarion Verlag 
edition, Munich, 1920-1928), xiv, p. 75 ; xv, p. 223 ; xvi, p. 224; 
XVII, p. 136; et passim. Those who are strong, Nietzsche de- 
clares, strive away from one another with as natural a necessity 
as the weak strive toward one another. Such passion of distance 
with respect to others, he adds, is the basis on which to promote 
"that yet more mysterious passion — that craving for more and 
more extension of distance within one's own soul," which leads 
toward that ever greater tension which is the "self -overcoming 
of man." 

An attempt to regard Heraclitus as a pessimist in the manner 
of Schopenhauer was made, for instance, by Gottfried Mayer (see 

12. The practice of dedicating a book in a temple was not 
unusual. Plutarch tells of a poetess who, having won a prize 
at the Isthmia for her poetry, reverently deposited a scroll of 
it in the temple at Delphi. Moreover someone, whether the poet 
himself or another, had dedicated the poems of Hesiod at Mount 

[ "5 ] 


Helicon, where Pausanias reports having seen them engraved 
on ancient tablets of lead. Nor was the dedicatory custom con- 
fined to poems. Oenopides of Chios dedicated a bronze copy 
of an astronomical table at Olympia, Xenocrates dedicated at 
a shrine on Mount Olympus his calculation of the height of 
the mountain, Eudoxus dedicated his astronomy at Delos, and 
a Carthaginian traveler named Hanno dedicated the log book of 
his travels in the temple of Baal at Carthage. These and other 
instances are collected by W. H. D. Rouse in The Votive Offer- 
ings (Cambridge University Press, 1902, p. 64). Cf. Pausanias 
IX. 31.4. 

13. In the pseudo-Aristotelian treatise, De Mundo (v:396 b, 
20) Heraclitus is called **the obscure" (6 aKOTetvo^) ; Diogenes 
Laertius quotes a saying that he is "riddling" (atviKTrj?) ; Ter- 
tullian speaks of him with apparent contempt as ille tenebrosus; 
while Clement of Alexandria declares of him that "he loved to 
conceal his metaphysics in the language of the Mysteries," and 
the remark has been widely quoted. Aristotle (Rhetoric iii. v : 
1407 b, 13-15) says: "To punctuate Heraclitus is no easy task, 
because we cannot tell whether a particular word belongs to 
what precedes or to what follows." Whatever truth there may 
be in these charges, the main reason for the obscurity and dif- 
ficulty in Heraclitus was not anything so simple and naive as a 
wish to mystify, but rather a need to speak appropriately and 
not too inadequately of that Nature which surprises us with 
the unexpected, which does not affirm or deny but merely gives 
signs, and which loves to hide (Frs. 19, 18, 17 respectively). 

14. "Lastly, there are Idols which have immigrated into 
men's minds from the various dogmas of philosophies, and also 
from wrong laws of demonstration. These I call Idols of the 
Theatre; because in my judgment all the received systems are 
but so many stage-plays, representing worlds of their own cre- 
ation after an unreal and scenic fashion. Nor is it only of the 
systems now in vogue, or only of the ancient sects and philos- 
ophies, that I speak; for many more plays of the same kind 
may yet be composed and in like artificial manner set forth; 
seeing that errors the most widely different have nevertheless 
causes for the most part alike. Neither again do I mean this 
only of entire systems, but also of many principles and axioms 
in science, which by tradition, credulity and negligence have 
come to be received." Francis Bacon, The Great Instauration 
(1620) : Part 11, "Which is called The New Organon; or, True 

[ 116 ] 


Directions Concerning the Interpretation of Nature," Sec. xliv. 
Ignoring the particular examples of the Fourth Idol which 
Bacon later gives by way of illustration, and which represent 
certain intellectual dangers of his day rather than of ours, we 
need to be vigilant in all of our thinking lest we fall into an 
analogous fault. In undertaking to study a distant thinker such 
as Heraclitus are we in danger of being hindered by "stage- 
plays, representing worlds of our own creation after an unreal 
and scenic fashion" ? I think we are. To discover our own Idola 
Theatri is much more difficult than to discover those of other 
persons and of remote civilizations and to label them after our 
own fashion. Nevertheless a clue to self-discovery in this re- 
spect is available for those who wish it. Everett Dean Martin 
once popularized the adage, "A man is known by the dilemmas 
he keeps." Each of us has certain preferred ways of setting 
up alternatives, hence of asking questions, or of setting his own 
interpretations upon questions that have already been raised. 
When any favorite pair of alternatives becomes so stereotyped 
that all our questions (however apparently free and far-ranging 
they may be) are asked in terms of that particular "either-or," 
or are asked as covertly presupposing it, that is equivalent to 
saying that an Idol of the Theater has grown up and taken 
possession, usually unconscious possession, of our minds. There 
are two such Idols of the Theater — two strongly held ways of 
presupposing what the possible types of reply are to a given 
question — which are current today and which, unless we can 
look beyond them, will be barriers to an adequate understanding 
of Heraclitus. They are: (i) the alternatives of subjective vs. 
objective, or mind vs. matter, or self vs. the world; and (2) 
the alternatives of a somehow loosely Christian way of thinking 
about religion vs. a rejection of the possibilities represented by 
religion altogether. The result is that two of ''the dilemmas that 
we keep" tend to take their shape from these Idols: we ask 
(i) whether such a thing as goodness, or beauty, or a heard 
sound, is in me (i.e., purely subjective and relative) or in the 
world itself (in the sense that houses and trees are there) ; and 
we ask (2) whether the Christian picture of God and the after- 
life is true or whether all religion is merely fanciful. Now from 
the point of view represented by these two ways of formulating 
questions (these two Idols of the Theater) Heraclitus will ap- 
pear to violate the law of contradiction with respect to the for- 
mer, since for him the Logos is both subjective and objective 

[ 117 ] 


at once (as, indeed, is the tao of the ancient Chinese, the dharma 
of Hinduism, and the holy pneuma of early Christianity), and 
he will appear to violate the law of excluded middle with re- 
spect to the latter, since his notion of divinity (cf. Frs. 6i, 62, 
63, 64, 74, 104, 105, 106, and 119) and his notion of the after- 
life (cf. Frs. 59, 65, 67, 68) open up possibilities that are not 
included in either of our familiar post-Christian alternatives. 

15. In his Conversations with Goethe, under the date Octo- 
ber 29, 1823, Eckermann quotes Goethe as saying: "Every 
character, however peculiar it may be, and every representation, 
from stone all the way up the scale to man, has a certain uni- 
versality; for everything repeats itself, and there is nothing in 
the world that has happened only once." Fritz Strich, com- 
menting on this and similar passages, remarks: "The symbol 
is thus, in Goethe's sense, the fullest coalescence of the particu- 
lar instance and a general idea" (Der Dichter und die Zeit, 
Bern, 1947). Likewise Coleridge praises Shakespeare for effect- 
ing a "union and interpenetration of the universal and the par- 
ticular" {Lectures on Shakespeare) . A part of the aim of au- 
thentic poetry is to unbar the representations of logical language 
and categorizing thought by rehabilitating certain earlier models 
of perceiving, thinking, and saying. 

16. Aristotle (De Anima iii. ii) on the relation of perception 
to what is perceived, and Chap, iv, on the relation of mind to 
that which is intellectually apprehended. 

17. The phrase was used by John Keats in a letter to his 
brother George, December 181 7, apropos of King Lear: "It 
struck me what quality went to form a man of achievement, 
especially in literature, and which Shakespeare possesses so 
enormously. I mean negative capability; that is, when a man 
is capable of being in uncertainties, mysteries, doubts, without 
any irritable reaching after fact and reason." 


1. Sextus (cf. note to Fr. i in Appendix B) calls Fr. i a 
"prefatory statement" {irpoeip-qixivov) and says that it gives a 
clue to the scope (irepiixov) of Heraclitus' treatise on nature. 

2. Burnet writes : "The Aoyo? is primarily the discourse of 
Herakleitos [sic] himself; though, as he is a prophet, we may 
call it his 'Word.' " For the remainder of his argument see 
Burnet, p. 133, n. i. Cf. note to Fr. 118 (Appendix B). 

[ 118 ] 


3. Wilamowitz in his edition of Euripides' Herakles adduces 
evidence to support his view that the more archaic Greek writers 
did not employ titles descriptive of the contents of the work, 
but simply began with the declaration, "So-and-so speaks as 
follows" (Aeyet raSe). The Pythagorean physician Alcmaeon, 
roughly contemporary with Heraclitus, opens his treatise with 
the words: "These are the words of Alcmaeon of Croton, son 
of Pirithous, which he spoke to Brontinus, Leon, and Bathyl- 
lus. . . ." However, since Alcmaeon lived in Italy, his practice 
offers no strong evidence regarding Heraclitus. At any rate, 
a number of scholars subscribe to Diels* opinion that Heracli- 
tus' treatise probably began in some such manner. 

4. "The common and divine Logos" : rov kolvov Xoyov koI Oelov. 
"By participation in which" : ov Kara ixeroyr^v. Sextus Empiricus, 
loc. cit. (in Note i). Sec. 131. It might be objected that I 
prejudice the argument by capitalizing "Logos" (Aoyo^) ; but it 
is equally possible that the use of a small initial letter might 
prejudice the argument in the contrary direction. I employ a 
capital only when, as here, the context makes it evident that 
the primary reference is cosmological. 

5. The view that what is perceived by many observers is 
trustworthy — ^provided they do not have "barbarian souls" (Fr. 
13) — and that what is perceived unsharably by a single ob- 
server is false, represents an important postulate upon which 
the structures of physical science have been erected. The scien- 
tific method that had been developed at Miletus involved a 
greater emphasis upon shared observation than was to be found 
in the visionary wisdom of poets and "wise men" (<to<I>ol). 
But in the few extant fragments of the Milesian philosophers 
there is no evidence that they had endeavored to formulate the 
nature of the newly objective method that they were beginning 
to employ. Heraclitus was more explicitly aware of the problem, 
as Frs. i, 2, 3, 5, 11, and 15 especially indicate. No doubt the 
widely shared type of experience that is employed in science 
needs to be, and inevitably is, supplemented by some degree of 
private insight as well; and surely no reader of Heraclitus will 
say that such insight is lacking in his utterances. Nevertheless, 
it was of great importance in the early development of science 
to have the claims of objective observation recognized; and 
Heraclitus contributed signally to such recognition. 

6. Epicharmus, fr. 57 (Diels-Kranz, p. 208). In contrasting 

[ "9 ] 


the divine Aoyo? with human calculation Epicharmus employs 
the word Aoyio-/>tos for the latter idea. 

7. Kirk and Raven, p. 188. Most of what Kirk and Raven 
write on the subject is excellent, but their use of the word "for- 
mula" as applied to the Logos strikes me as gravely misleading. 

8. Thus Heraclitus is enabled to say in Fr. 81 that men 
should speak with rational awareness {^v v^) and thereby hold 
on strongly to that which is shared in common (tw |ww). Cf. 
Notes to Frs. 2 and 81 (Appendix B). 


1. I follow the custom of translating yeVeo-t? and <l>Oopd by 
the compounds ''coming-to-be" and "passing-away." Simpler 
pairs of words, such as "creation" and "destruction," or "ap- 
pearance" and "disappearance," are likely to carry misleading 

2. Aristotle Physics i. vi: 189 a, 22ff. Aristotle adds: "The 
same difficulties hold for every other pair of opposites: Love 
is not to be thought of [vs. Empedocles' theory] as gathering 
up Strife and creating something out of it, nor can Strife do 
this to Love, but rather both of them must operate on a third 
something." His illustrations are designed to support his con- 
tention, against Heraclitus, that an adequate explanation of the 
world requires more than the principle of duality (opposition, 
strife), that it requires also the principle of triadicity, in the 
sense that any adequate conception of change involves three 
notions — of a prior state, a posterior state, and a "something that 
remains unchanged throughout the process." 

3. "War and Zeus are the same thing" : tov iroXefiov kol tov 

Ala TOV avTOv ctvat, Kaddirep kol tov ^HpaKAeirov Aeyeiv. Philodemus, 

in Dox., p. 548, with a few missing letters restored by Diels. 

4. On destiny and necessity : 'HpaKAetro? iravTa Ka9' elfiapfxev-qv, 
Tr]v Se avTrjv virapxeiv dvdyKrjv. Stobaeus and Plutarch, in Dox., 
p. 322. Cf. Theophrastus, who says of Heraclitus that "he posits 
a certain order and a definite time in which cosmic change comes 
about in accordance with a certain destined necessity" (/cara nm 

elixapfievrjv dvdyKrjv) — DoX., p. 47^' 

5. Aristotle Physics 11. iv-vi : 195 b, 31-197 a, 13, on chance; 
and II. ix: 199 b, 34-200 b, 9 on necessity. Just how Aristotle 
intends to relate these two concepts is a controversial question; 
but he seems to me to regard them as alike in that they represent, 

[ 120 ] 


in different perspectives, the kind of occurrence or the aspect of 
occurrence that is independent of intelHgent purpose. A chance 
event can be defined (so I interpret 198 a, 5-7) as an unpur- 
posed and hence "necessary" event which happens to affect some 
human interest favorably or unfavorably, so that it looks as if 
it had occurred for the sake of that human interest or in order to 
thwart it, although in fact it did not. 


1. Oswald Spengler takes fire as a symbol of iravra pel, which 
he interprets as the "formal principle" of physical nature (Der 
metaphysische Grundgedanke der herakleitischen Philosophic, A. 
I. 3). This he calls "the first formulation" (die erste Formu- 
lierung) of Heraclitus' doctrine, as distinguished from "the sec- 
ond formulation," the principle of paradox. 

2. Gustave Teichmiiller, Neue Studien zur Geschichte der 
Begriffe, Heft i, p. 2 et passim. The tendency to regard Hera- 
clitus as an Ionian philosopher carrying on the tradition of 
Miletus, and hence to take his concept of fire in a purely physical 
sense, is an oversimplification that is found in a number of 
writers (e. g., Julio Navarro Monzo, La Busqueda Presocrdtica) 
and has unfortunately become lodged in several textbooks. 

3. Aristotle De Caelo iii. v : 303 b, 12. 

4. Aristotle Metaphysics i. iii : 984 a, 7. 

5. The main Greek commentators on Aristotle quite generally 
take the Heraclitean concept of fire in a material sense. Asclepius 
repeatedly refers to Heraclitus' fire in a context which shows 
that he considers it as playing a role analogous to that of water 
in Thales' system and of air in the system of Anaximenes, which 
he describes by the phrase, etvat irpwrov oltlov v\lk6v {Cont- 
mentaria in Aristotelcm Gracca vi, Pt. 11, p. 25 ; cf. p. 148, line 
19). Elsewhere Asclepius says of Heraclitus and the Milesians 
jointly that each one of them took one of the physical elements 
(in Heraclitus' case, fire) to be "a bodily and material principle" 

(GroifiaTLKrjv OLpyrjV €7roiovv kol v\lkt]v) . 

Alexander Aphrodisiensis indicates a similar view in his com- 
mentary on Aristotle's Metaphysics, in Comm. Ar. Gr. i, p. 45. 
Later, however, (op.cit., p. 670) he speaks of the Heraclitean 
fire as being not only apxrf but also ovma. The use of the latter 
word suggests that he may have had something more than a 
physical principle in mind, but without making clear what it is. 

Simplicius, in Comm. Ar. Gr. vii, p. 621, says that Heraclitus 

[ 121 ] 

regarded fire as the "primary bodily principle" {irpdrov twv 


6. **Fire for him [Heraclitus] was neither a mere symbol of 
the universal process nor a substrate persisting as identical 
throughout its qualitative alterations. He speaks of it both as a 
token for exchange like gold in trade and as involved in change 
itself; and it was the easier for him in this case to identify the 
sign and the thing signified, since fire does appear to be the one 
existing phenomenon that is nothing but change." — Harold 
Cherniss, 'The Characteristics and Effects of Presocratic 
Philosophy," Journal of the History of Ideas, xii (1951), p. 331. 

7- Hippolytus' words are : Aeyet 8e koI cfypovifiov tovto ehab TO 

TTvp KOL T-fj<s StoiKT^crew? T(ov 6Xo)v aiTLov. Legge's translation runs : 
"But he also says that this fire is discerning and the cause of the 
government of the universe." Thereupon follows the statement 
which is Fr. 30 : "And he calls it craving and satiety." 

Stobaeus' words are : H/oa/cAetro? ov Kara xpovov ehat yevrjTOv rbv 
Koa-fiov, dAAot Kar inLvotav (Dox., p. 331). A translation is given in 
the text. Kirk (p. 356) suggests that Heraclitus may have con- 
sidered not all fire to be rational, but only fire of the purest and 
most ethereal sort. 

8. The seventeenth century chemist Robert Boyle speaks of 
"those two most grand and most catholic principles of bodies, 
matter and motion" (Works, iv, p. ^2, "On the Excellence of 
the Mechanical Principle"). The word "motion" in Boyle's con- 
text means movement in space, not qualitative change. It is 
therefore both more exact and semantically more restricted than 
the Greek word dv-qai^. The acceptance of spatial movement as 
ontologically prior to other kinds of change is so deeply in- 
grained in our contemporary thinking that we must make a 
deliberate effort to realize that the Greeks for the most part did 
not share this presupposition; and certainly Heraclitus did not. 
Cf., by contrast, W. A. Heidel, "Qualitative Change in Pre- 
Socratic Philosophy," in Archiv zur Geschichte der Philosophie, 

19 (1906), pp. 333-379. 

9. Aristotle {De Caelo iir. v: 304 a, loff.). The analogy on 
which the second argument is based, he says, is this : "The finest 
body is fire, while among figures the pyramid is primary and has 
the smallest parts." Simplicius' comment, denying that Heraclitus 
conceived of fire in terms of pyramids is to be found in Com- 
mentaria in Aristotelem Graeca, vii, p. 621, lines 7ff. 

10. Epicurus' reference to the nprjcrTrip is in his letter to 
Pythocles, Sees. 104-105 : Epicurus, The Extant Remains, ed. 

[ 122 ] 


Cyril Bailey (Oxford, 1926), p. 71. In Lucretius the word 
7rpr}(TT7]p occurs in Book VI, line 424; his meteorological descrip- 
tion in lines 423-450 is relevant to the subject. Cyril Bailey's 
quoted remark occurs in his commentary on line 424, in Vol. 
Ill, p. 1618, of his edition of Lucretius (Oxford, 1947). Seneca's 
reference to the same phenomenon is in his Quaestiones Naturales 
V. xiii. 3. Cf. Burnet, pp. 149-150; Kirk, pp. 325-331. 

11. Kirk, p. 356. Plutarch evidently considers the Trp-qarrip and 
the Kepavvo^ as phenomena of much the same type : Dox., p. 275, 
line 2. 

12. Aetius-Plutarch, in Dox., p. 276. 

13. *'We come into existence by an efflux from air" : Kar 

eKTTOiav rovTov ytyvo/xe^a, where the WOrd rovTOV refers to a-qp in 

the preceding clause: Anaximenes, fr. 3 in Diels-Kranz. Al- 
though the fragment is repudiated in Diels-Kranz as "gefdl- 
schtes" — i.e., as probably not the actual words employed by 
Anaximenes — this possibility does not affect the point made in 
the chapter. 

14. Kirk, p. 115, declares of Fr. 34 that "the presence of air 
shows that we are dealing with a Stoicized version of Heraclitus." 
While this may be so, my argument in the chapter is intended to 
show another possibility. 

15. Plutarch, On the E at Delphi, 392 C. 

16. Galenus, Historia Philosophiae, in Dox., p. 626. 

17. Both passages mentioning the alleged theory of Hera- 
clitus that air is primary are to be found in Sextus Empiricus, 
Against the Physicists (Vol. iii of the Loeb Classical Library 
edition). The former passage is in Bk. i. 360 {ibid., pp. 173, 
175) ; where my phrase "the primary and fundamental elements" 
is a translation of apxi] koI o-rotxetov. The latter passage, declaring 
on the authority of Aenesidemus that Heraclitus held the primary 
existent (t6 6v) to be air, is in Bk. 11, Sec. 233 (ibid., p. 325) of 
the same essay. 

18. Cf. Mircea Eliade, The Myth of the Eternal Return 
(Bollingen Series, xlv; Pantheon Books, 1954), particularly 
Chapter 11, "The Regeneration of Time." 

19. Kirk, pp. 336-338. Although Kirk's fifth point seems to 
comprise two distinct arguments, I have followed his numbering. 
His pages 307-335, involving interpretations of Frs. 29, 32, and 
33 (corresponding to Diels 30 and 31) are indirectly relevant. 

20. Plato Sophist 242 D-E. 

21. See, for instance, the Theaetetus 152 E, f¥., where the 
Platonic Socrates is attacking the doctrine, evidently still in- 

[ 123 ] 


fluential, that all things that we say "exist" are really in process 
of becoming — a doctrine that he ascribes loosely to a "whole 
series of philosophers," including Protagoras, Heraclitus, Empe- 
docles, and even Homer. 

22. Aristotle De Caelo i. x : 279 b, 15-17. J. L. Stocks' transla- 
tion (Oxford, 1922) is used here. Stocks interprets Heraclitus 
as believing "in periodic changes in the constitution of the world 
as a whole" and that "the world exists, as it were, in a succession 
of lives." He adds, reasonably I think, that Aristotle's phrase 
"alternation in the destructive process" is somewhat misleading, 
since the supposed alternation is rather between generation and 
destruction — ^an antithesis which Stock finds analogous to the 
Love and Strife of Empedocles. Simplicius, in commenting on 
the passage in De Caelo, employs the verbs iKirvpovadai and 
(TvviGTaaOai to signify respectively the fiery dissolution of the 
universe in a general conflagration and the subsequent building 
up of a new universe. (Commentaria in Aristotelem Graeca vii, 
p. 294.) 

In an epistle of Plutarch and in Stobaeus' Eclogues (com- 
pared in parallel columns by Diels, in Dox., pp. 283-284) 
Heraclitus is quoted as saying : "Fire is the first principle of all 
things; for all things arise from and eventually pass back into 
fire." The statement is omitted from the present collection of 
Fragments, because its authenticity has been challenged by Diels, 
By water, and a majority of classical scholars. Plutarch, whose 
version is somewhat fuller than that of Stobaeus, goes on to 
elaborate the intended quotation with the statements that the 
universe is created through the quenching of the fire, and that 
after the lengthy process of cosmic creation (what today we call 
evolution) has been completed, the universe and all the bodies 

in it (tov Koa-fiov Kal iravra tcl a-Mfiara) are once more (iraAtv Se) 

dissolved by fire in the general conflagration (iv Tfj cKTrv/awo-et). 
Since Plutarch employs indirect discourse in making these state- 
ments, he evidently intends them (whether rightly or mistakenly) 
as quotations from Heraclitus. Granted that some of the language 
is probably of Stoic origin, there is no reason why the ideas 
(which are easily envisaged in clear imagery) may not be much 

23. Aristotle Physics iii. v : 204b, 35 - 205a, 4. The significant 

words are : aSvvaTov rb irav . . . rj ctvat rj ytyvecrOaL ev tl avTwv. 
wCTTTcp ^HpaKAeiTO? cf)r}(nv arravra yiyveo-Oat ttote irvp. 

On the other hand Cherniss argues that the subject of the 

[ 124 ] 


clause referring to Heraclitus is not avavra but nvp. According to 
this interpretation Heraclitus would be saying not that all things 
become fire, but that fire becomes all things; hence Aristotle 
would be citing him (although Cherniss does not say so) as an 
ally in this instance, instead of as an opponent. That is, he would 
be drawing upon Heraclitus for confirmation, as though he were 
to say : "Fire cannot be infinite, existing alone and in the absence 
of other real things; and indeed Heraclitus agrees, for he says 
that his fire is constantly becoming other things than itself." 
See Harold Cherniss, Aristotle's Criticism of Presocratic Phi- 
losophy, p. 29, n. 108. But it would be a surprising departure 
from Aristotle's usual attitude if he were to call on Heraclitus 
for confirmation! 

Taking, then, airavra to be the subject and irvp to be the object 
of the infinitive verb yiyveaOou (as most interpreters other than 
Cherniss have done), the one technical question that remains is 
whether aTravra is to be understood dissociatively or collectively. 
That is, does the clause assert that all things are constantly in 
process of becoming fire (which is the likeliest meaning of Fr. 
28) or that at certain times (ttotc) everything becomes fire 
simultaneously ? Zeller is of the opinion that airavra connotes the 
simultaneous totality of all things, as distinguished from iravra 
which would connote all things discursively and individually. 
Burnet, on the contrary, denies that there is any appreciable 
connotative difference between the two expressions. It appears 
that the debated question cannot be answered by appealing to this 
word alone. 

24. The statement in Aetius is followed by an explanation, 
which may be intended as a paraphrase of Heraclitus' views ; it 
is impossible to be sure. The passage runs as follows : "Through 
the quenching of this fire all the things in the universe are made. 
First, what is coarsest is drawn of¥ into itself, producing earth ; 
then the earth, through the loosening action of fire, turns into 
water ; the water is evaporated into air ; and then once more the 
universe and all the bodies in it are dissolved by fire in the 
general conflagration again" (Dox., p. 284). 

25. The Pseudo-Aristotelian De Mundo 401 a, 8f¥. : "All 
animals, both wild and tame, feeding in the air or on the earth 
or in the water, are born and mature and decay in obedience to 
the ordinance of God; for in the words of Heraclitus . . ." — 
whereupon the Fragment is quoted. 

[ 125 ] 


1. Arius Didymus, in Dox., p. 471, line 2. 

2. Aristotle De Anima i. i: 403 a, 24: the affections (m 
TrdSr), where "of soul" or "of the soul" is understood) are mean- 
ings implicit in matter (Aoyoi evvXot). Later, in Bk. 11 of De 
Anima, Aristotle defines soul itself as "the first grade of actuality 
of a natural being having life potentially in it" (412 a, 20-21 : 
J. A. Smith's translation in the Oxford University Press series) 
and as "the first grade of actuality of a natural organized body" 
(412 b, 5-6: the same). 

3. Theophrastus, in De Vertigine, mentions eyesight in par- 
ticular, for if the motions of the head should stop, or be deranged, 
the clarity of the visual image would fail. The nature of the 
image, he observes, is preserved and exists (o-w^erat koX o-u/x/xeVct) 
through this very motion (/ctvTyo-ts). See Walzer, p. 154. 

4. Kirk writes: "Souls use smell in Hades because they are 
surrounded by dry matter, than which they are but little less dry. 
When one recalls that the soul in life was by implication charac- 
terized as a form of fire, it is not difficult to deduce that Hera- 
clitus' 'Hades' is a realm of fire, in which the disembodied souls 
are themselves fiery." (G. S. Kirk, in American Journal of 
Philology, 70 [1947], p. 389.) Cf. Appendix B, the Note to Fr. 


1. Euripides, The Trojan Women, line 885. Sextus Empiricus 
quotes the passage as preliminary to a discussion of the Hera- 
clitean Logos. 

2. Sextus Empiricus Against the Logicians i. 129-130. "The 
divine Logos" is 6 6do<: Aoyos. "Intelligent" : vo€.t6<5. "Mindful 
again" : ttciAiv ifx^poves. 

3. Xenophanes, fr. 15 in Diels-Kranz. 

4. Aristotle's mention of Thales' remark that soul is diffused 
through the entire universe is in De Anima i. 5: 411 a, 7-8. 
Simplicius' commentary is in Commentaria in Aristotelem 
Graeca, xi, p. 73. Thales' statement about the magnet having 
soul is mentioned by Aristotle in De Anima 405 a, 19. 

5. Cebes' argument, that the relation of soul to body may be 
analogous to the relation of body to its clothes — the soul out- 
living a number of bodies, as the body outlasts a number of pieces 
of clothing, although neither of them lasts eternally for all that 

[ 126 ] 


— is given in Plato's Phaedo 87A - 88C. Socrates' reply, such as 
it is, appears in 95 A-E. 

6. My discussion of plurisignation is in The Burning Fountain 
(Indiana University Press, 1954), pp. 61-62, 112-117, 149-151, 
and, with reference to ancient Greek thinking, pp. 254-255. 

7. G. S. Kirk, ''Heraclitus and Death in Battle," in American 
Journal of Philology, 70 (1949), p. 390. Cf. Appendix B, the 
Note to Fr. 86. 


1. See Note 6 to Chapter viii. I am not, of course, attributing 
to Aristotle the extended significance of aixaprU suggested in 
the text. It may be, as S. H. Butcher says {Aristotle' s Theory 
of Poetry and Fine Arts, p. 295) that Aristotle used the word 
in much the same sense as he normally used afxaprrjixa, which 
he defines (Nicomachean Ethics v. viii 11135 b 16) as a mistake 
that could have been avoided. Butcher adds, however, that both 
words are also occasionally used by Aristotle in the broader 
sense of error due to unavoidable ignorance, as virtually synony- 
mous with aTvxqy^a, '^misfortune." Now, although in the Ethics 
Aristotle maintains a careful distinction between actions done 
with the power of knowing and actions done in unavoidable 
ignorance, he says in the Poetics (xiv:i453 b 27-31) that the 
central action of a tragic plot can be of either type, and he men- 
tions the Oedipus Tyrannus of Sophocles as an example of the 
latter. Obviously, of course, Oedipus acted in ignorance of what 
he was doing ; yet Sophocles calls him, in the title, rvpawo^, which 
carries the connotation of usurper. Oedipus, then, has usurped; 
and much of the power of Sophocles' play comes from the im- 
plicit sense that a potentially universal human condition is rep- 
resented by the tragic action. All men are usurpers in some 
sense ; all things, by their very existence, usurp ; and such usur- 
pation is, in the broadest sense, a kind of hybris. 

2. Cf. Geoffrey S. Kirk, "Heraclitus and Death in Battle," 
where possible meanings of Fr. 86 are discussed. 

3. On the semantic role of the serious pun, see William 
Empson, Seven Types of Ambiguity (Oxford University Press, 
1930) ; and cf. the references in Chapter v. Note 6. 

4. Kirk, pp. 50-51. As against Reinhardt, Parmenides, pp. 
2i5if., and Gigon, Untersuchungen zu Heraklit, p. 14. The point 
is not that there was no difference to the ancient mind between 

[ 127 ] 


codified laws and unwritten laws — ^for of course there was, and 
such unwritten laws as those bearing on one's duties to the gods, 
to one's parents, and to guests were regarded as the more funda- 
mental — but that both these types of law were ol avOpMiraoi vofioi, 
and as such were secondary to and nourished by the one divine 

law : VTTO evo<; tov Odov. 

5. The episode of Byrsa and the bull's hide is told in the 
Aeneid i. 365-370. The episode of the wooden horse is told in 
II 234ff. Line 234 itself is of special importance: Dividimus 
muros et moenia pandimus urbis. In order to make the gate large 
enough for the wooden horse to be dragged through it, the 
Trojans had to chop asunder a part of their city wall and thereby 
"open the defenses of the city." Apart from the unknown threat 
inside the horse, this weakening of the city wall was bad enough 
from a mihtary point of view, with the enemy lurking a little 
way off. It was even worse, to the ancient mind, in its symboli- 
cal and magical aspect. 


1. Asclepius, commentary on Aristotle's Metaphysics, in 
Commentaria in Aristotelem Graeca, vi, Pt. 11, p. 258. 

2. I have developed the distinction between depth language 
and steno-language in The Burning Fountain, pp. 24-29; 48-51. 

3. Hippolytus The Refutation of All Heresies ix. He describes 
the motion of the screw by saying, "For the upward way is en- 
compassed by [or cleaves to ?] a circle [or circular motion]": 

ai^o) yap ofxov Kal kvkXo) Trepte-^eTai. 

4. Alcmaeon of Crotona, Fragment 2 in Diels-Kranz. He is 
believed to have lived about a generation after HeracHtus. Little 
else is known about him, and there are only four other surviving 
fragments of his writing. 


1. Hastings' Encyclopaedia of Religion and Ethics. After the 
words, "truth and illusion," Diels inserts the corresponding 
Greek words that he has in mind, aXrjOeia and S6ia. 

2. Hermann Frankel, "Heraclitus on God and the Phenom- 
enal World" : American Philological Association, Proceedings, 
69 (1938), pp. 230flf. 

3. Phaedo 85 E, ff . Simmias, a Pythagorean, uses the word 

[ 128 ] 


apixovla as connoting ''right proportion among such bodily ele- 
ments as hot and cold, moist and dry." If this were all, Kirk's 
nonmusical interpretation of the word might appear to be justi- 
fied. But in the previous sentence Simmias has been speaking of 
the dpfiovia produced by the strings of a lyre; from which the 
nature of his metaphoric image is evident. 

4. Aristotle (Nicomachean Ethics viii. 1:1155 b, 5) ostensibly 
quotes Heraclitus as saying that the fairest apfxovia is produced 
from opposing tones {Ik twv Siacf)€p6vT(Dv, where the word for 
**tones" is evidently to be understood). Aristotle connects this 
with the statements, which he also attributes to Heraclitus, that 
opposites unite and that all things occur through strife. Diels- 
Kranz does not accept these remarks as authentic Heraclitean 
quotations, however. 

5. Edmund Pfleiderer, Die Philosophie des Heraklit von 
Ephesus im Lichte der Mysterienidee, p. 89. 

6. Aristotle Poetics xiii:i453 a, 10, where effective tragedy 
is said to occur 81 aixaprlav TLvd. The cognate verb dfiaproveiv can 
be seen developing its metaphorical meaning in Homer; for in 
the Iliad (v. 287) it is applied to a spear's missing its mark, and 
by the time of the Odyssey (xxi. 155) it has come to mean, 
"'to fail of one's purpose." 

7. Marcus Aurelius Meditations vi. 42. The passage in which 
the quotation occurs runs as follows, in Long's translation: 
"We are all working together to one end, some with knowledge 
and design, and others without knowing what they do; as men 
also when they are asleep, of whom it was Heraclitus, I think, 
who says that they are laborers and cooperators in the things 
that take place in the universe. But men cooperate after different 
fashions; and even those cooperate abundantly who find fault 
with what happens and those who try to oppose it and to hinder 
it ; for the universe has need even of such men as these." 

[ 129 ] 


Since the philosophy of Heraclitus is available to us only 
through quotations of him and references to him made by later 
ancient writers, an editor attempting to reconstruct the original 
doctrine faces two kinds of problem. The one, the problem of 
interpretation, arises in connection with any philosophy that may 
be studied, although the problem is likely to be more acute when 
the philosophy survives only in a fragmentary state. The other 
problem, which arises when a philosopher's utterances are known 
only through quotations by others, has to do with authenticity. 
My own judgments of authenticity are represented by the one 
hundred and twenty-four Fragments contained in the present 
list — not counting the authentic but nugatory bits that I have 
relegated to Appendix C. Although in the main my judgments 
of authenticity have been guided by those of Diels (fifth edition, 
as revised by Kranz, 1934), it will be found that I have in- 
cluded five Fragments (20, 27, 43, 62, 74) that were omitted 
from the Diels-Kranz list. My reasons for accepting them are 
stated in each case. 

One's interpretation of a particular Fragment, as of a word 
within a Fragment, often requires (besides a dictionary knowl- 
edge of the meaning of the words employed) two kinds of criti- 
cal examination, which must sometimes be balanced against each 
other. The meaning of a quotation, wherever it is in doubt, 
must be studied in relation both to the doxographical context 
and to whatever other relevant evidences there are concerning 
Heraclitus' views on the subject involved. On the one hand 
we must try to be critically aware of how and why a certain 
writer introduces a quotation from Heraclitus, and must con- 
sider whether the writer's general attitude and purpose might 
be such as to tempt him to give the quotation an emphasis or 
a twist that may be foreign to the original intention. On the 
other hand, we sometimes have to raise the question, "Does 
this or that interpretation of word or sentence fit in with what 
we judge, from other Heraclitean quotations and to some extent 
from peripheral evidence, that Heraclitus most probably would 
have meant?" This sort of criterion is an especially risky one 
when applied to a philosophy so full of paradoxes as that of 
Heraclitus; nevertheless it cannot be altogether avoided. 

[ 130 ] 


Suppose, for example, we are considering Clement of Alexan- 
dria's quotation of Fr. 19, which is here translated, "Unless you 
expect the unexpected you will never find [truth] ; for it is 
hard to discover and hard to attain." A lively dispute has raged 
over the question of whether there should or should not be a 
grammatical break after the third word, eXvrjTaL. If the tradi- 
tional reading is accepted, as given by Diels, retaining the comma 
after the third word, then the translations of Fairbanks (repub- 
lished in Nahm's well-known volume of pre-Socratic philos- 
ophers) and of Kathleen Freeman are appropriate: **If one does 
not hope, one will not find the un-hoped for" (Freeman). If, 
on the other hand, Gomperz's suggested emendation of the tra- 
ditional reading is accepted and the grammatical break is delayed 
until after the fourth word, aveX-TnaTov, then the latter word be- 
comes the object of the foregoing verb, and the meaning be- 
comes, as expressed by Burnet: "If you do not expect the un- 
expected, you will not find it; for it is hard to be sought out 
and difficult." (The antecedent of "it" has been lost; I assume 
that it expressed in some manner the general but elusive object 
of one's highest search.) Now unquestionably the Diels punctua- 
tion, and not the Gomperz emendation, is in line with the way 
in which Clement himself understood and reported the passage, 
as shown both by the manuscript reading and by Clement's 
context. But since probably no sign for the comma-break had 
been in use when Heraclitus wrote, and since conventions of 
comma-usage were still fitful in Clement's day, it is easy to see 
why Clement might in good faith have shifted the position of 
the break from where Heraclitus had intended it should be. 
In such a situation our examination of the doxographical con- 
text can bring to light some of the ideas that might have been 
present in Clement's mind to produce such an unconscious alter- 
ation of the original meaning. Now if Clement's misinterpreta- 
tion is to be taken not as an abstract possibility but as a sig- 
nificant hypothesis, then we must look back also to whatever we 
can soundly infer regarding the original context — I mean, of 
course, not the original verbal context, of which we know noth- 
ing, but the original philosophical context, consisting of Hera- 
clitus' characteristic ways of thinking and expressing himself — 
in order to have a basis for judging how the passage was most 
probably meant when it was first written. 

In the case of the present Fragment, an examination of the 
doxographical context shows that Clement's purpose in intro- 

[ 131 ] 


ducing the quotation was to let it stand parallel to his immedi- 
ately preceding quotation from Isaiah, vii. 9, which in the Re- 
vised Standard Version reads: '*If you will not believe, surely 
you shall not be established." The word that is translated "be- 
lieve" in the modern Bibles appeared as the verb Tnarevav in 
the Greek text that Clement was using; his quotation runs, 
lav fjLT} TTLCTTevarjTe, ovSe fxr] cTvvrjTe. Thus it seems evident that 
Clement was taking the verbs TrKTrevetv and eATrt^eiv as respectively 
connoting the ideas of faith and hope, which St. Paul had 
memorably united in the texture of Christian thought. On rec- 
ognizing this contextual tie-up in Clement's Stromata our minds 
naturally pass, as the next critical step, to the question: "But 
would Heraclitus' own doctrine have contained any such con- 
nection of ideas?" Obviously the only possible answer here is 
"No" ; for while we cannot be sure of what Heraclitus' original 
context of the remark may have been, we can be sure of some 
of the things it was not, and the Clementine Christian associa- 
tion of ideas is one of them. As a matter of fact, I believe we 
can go even a step further, and affirm that the paradox "expect 
the unexpected" appears, from what else we know of the Ephe- 
sian's style, to be rather distinctively and happily Heraclitean 
in tone, in a way that Clement's version is not. Accordingly it 
seems to me we have sufficiently objective grounds for rejecting 
the translations published by Fairbanks (1898), Nahm (1935, 
employing the Fairbanks translation), and Kathleen Freeman 
(1948), and for accepting those of Burnet (given above) and 
R. Walzer {''Se alcuno mai non speri Vinspearahile, non lo 
trover a''). 

A different sort of problem arises in the case of another 
Christian doxographer, Hippolytus. Like Clement he shows a 
marked Christian bias, but his bias does not appear to diminish 
his reliability when he is quoting from pagan sources, particu- 
larly from Heraclitus. This is the almost universal opinion of 
modern scholars; it is supported both by scattered evidence 
from comparison with other versions of a quoted Fragment of- 
fered by other writers, and also deductively from the nature 
and circumstances of his writing. The former kind of evidence 
can be discovered by comparing Fr. 66, taken from Hippolytus, 
with the variant form given by Clement; for as I argue in my 
note to the Fragment in the present Appendix, it is psychologi- 
cally more plausible to regard Hippolytus' as the original and 
Clement's as the variant version rather than the other way round. 

[ 132 ] 


To this type of specific evidence, which can be adduced in 
the case of several of the Fragments, there can be added a gen- 
eral consideration, drawn from the character of Hippolytus* 
thinking and composition. His purpose in writing The Refuta- 
tion of All Heresies he frankly avows. It is to demonstrate that 
the major Christian heresies up to his time had been produced 
not by a corruption of Christian faith from within, but by an 
infiltration of pagan ideas and beliefs. In the ninth book he deals 
with the heresy of Noetus, who had violated the orthodox creed 
by overstressing the identity of Father and Son ; and he under- 
takes to show that Noetus and his followers, "who delude them- 
selves into thinking they are disciples of Christ," are actually 
influenced by the teachings of Heraclitus and Stoicism. In carry- 
ing out this pious aim Hippolytus shows a strangely defective 
sense of logical connections. Choosing some of the most para- 
doxical of Heraclitus' utterances (a choice for which we may 
well be grateful) he throws them together in a hit-or-miss 
fashion, occasionally making farfetched comparisons with ele- 
ments of Christian doctrine. For instance, he introduces Fr. 68, 
*'They will arise into wakefulness and become guardians of the 
living and the dead," by explaining that Heraclitus is here 
affirming ''that there is a resurrection of this palpable flesh in 
which we are born." Fr. 25 ("War is both father and king of 
all . . .") and the latter half of Fr. 24 ("The royal power is 
a child's") he takes as distorted and proto-heretical references 
to, respectively, Father and Son in the Christian Trinity. Surely 
anyone enjoying so loose a sense of interpretative relevance as 
these passages evince would scarcely have reason for wishing to 
misquote ; for any quotations of a strikingly paradoxical charac- 
ter would serve his turn, and there must have been plenty of 
such material to be found in Heraclitus' treatise as it stood. It 
is Hippolytus' very lack of relational clarity that makes his in- 
dividual quotations the more reliable. 

Among the non-Christian writers, too, a critical reserve must 
guard our acceptance of alleged quotations. In the case of Plato 
and Aristotle, as I have argued elsewhere, there is reason to 
suspect that at times their references to Heraclitus are colored 
and perhaps exacerbated by the somewhat too provocative opin- 
ions of contemporary neo-Heracliteans. In the case of Diogenes 
Laertius the fault is likely to be a gossipy irresponsibility, and 
a scholar has the responsibility of judging which of his remarks 
are likely to have been affected by it and which not. In the cases 

[ ^33 ] 


of Plutarch and Sextus Empiricus it is not always easy to dis- 
cover the line between the quoted saying of Heraclitus and philo- 
sophical supplementations that come either from Stoic sources 
or from these writers themselves. Recognizing such tendentious 
characteristics of the various writers who quote Heraclitus is 
a necessary part of critical procedure. 

My readings of the Greek text of the Fragments have been 
guided mainly by Diels-Kranz and Walzer, supplemented wher- 
ever possible by the best available edition or editions of the 
doxographical sources: see Bibliography, Section i. Each Frag- 
ment is identified in the present Appendix by the number that 
introduced it in Chapters i to viii. In parentheses are given the 
numbers employed by Diels in his "B" list (adopted by Diels- 
Kranz, Walzer, Freeman, and virtually all German scholars of 
the present century) and by By water (adopted by Burnet, Fair- 
banks, and the Loeb Classical Library) . Words and phrases that 
the present editor judges to have had their source not in Hera- 
clitus but in the doxographers who quote him are enclosed in 
square brackets. 

A list of the more important doxographical sources of the Frag- 
ments is to be found on page 156. 


Fr. I (D I ; By 2). rov 8e Xoyov rov8i' e6vro<s act d^verot yivovTai 
avOpcoiroL Kal irpocrBev rj aKovdat Kal aKovo-avT€<s to tt/owtov* yLvofxevoyv 
yap iravTOiv Kara rov Xoyov rovhe aTrdpoia-iv loiKadt, TreipMfxevoL Kal 
eirioiv Kal epywv roLO'vroJV. 6kolo)V eyw StrjyevixaL Kara (f>vo'Lv Statpecov 
€Ka(TTov Kal cf>pd^(ov OK(j}<s e;)(€t* rov<s Be a\Xov<s avOpioTrov^ Xavddvet OKoaa 
iyepOevre^ TroLovaiv, oKUia-irep OKoaa evSovre^ liriXavBdvovrai. 

The passage is preserved by Sextus Empiricus, in Adversus 
Mathematicos ("Against the Savants"), Bk. vii, Sec. 132. In the 
Loeb Classical Library edition of Sextus Empiricus, Bk. vii of 
this treatise is published as Bk. i of "Against the Logicians," but 
the section numbers correspond. Sextus says that Heraclitus 
wrote these words "at the outset of his writings on nature" 
(ei/a/oxoju-evo? ovv rwv irepl cj>vo-e(ii<s) . Portions of the passage, but 
not the whole of it, are quoted by Aristotle {Rhetoric iii. 5), 
Clement of Alexandria (Stromata 11, p. 401, in Stahlin's edition), 
and Hippolytus (Refutation of All Heresies ix). 

There have been some disagreements as to how the opening 

[ 134 ] 


words should be divided into phrases. I follow the practice of 
Diels (as distinct from Diels-Kranz) and Burnet in taking det 
as attached to the words standing ahead of it. Kirk, who takes 
it as going with the words that follow, translates : *'0f the Logos 
which is as I describe it men always prove to be uncomprehend- 
ing . . ." (Kirk, p. 33; Kirk and Raven, p. i87n.). 

In order not to prejudge the meaning of the controversial 
word Aoyo9 I follow Kirk in not translating it but simply trans- 
literating it as "Logos." Diels translates it 'Wort (Weltgeset^)" ; 
Diels-Kranz, ''der Lehre Sinn." The question of the word's 
meaning is discussed in Chapter i. 

Fr. 2 (D 2; By 9^). Bl6 Set tireaOai Tw ^vv<a [rouTco-Tt tw koivw* 
^wos yap 6 Koivo?] . tov Xoyov S' iovro^ iwov ^movctlv 01 TroAAot w? 
ISlav €x^vT€^ (f>p6v7](nv. Sextus Empiricus, op.cit., Sec. 133. I 
follow Schleiermacher and Burnet in their reconstruction of 
the first sentence. This, and not the manuscript reading, makes 
logical connection with the remainder of the quotation. Bekker's 
theory is plausible: that the original quotation contained the 
Heraclitean word rw ^vw, that Sextus explained it to his readers 
by adding the phrase rovrian rw koivw, and that the words tw 
^vv<i> TovTecTTL wcrc droppcd out in copying. The next words, 
ivvos yap 6 Kotvo?, are almost certainly Sextus' gloss, as most 
scholars agree. Diels-Kranz prints these two sets of words after 
^vw as belonging to Sextus rather than to Heraclitus. Walzer, 
although he includes all the words in his text of the Fragment, 
limits his translation to the simple statement, "Bisogna percio 
seguire il commune." The word <f>p6vr)crLs, which I translate 
"intelligence," is translated "Einsichf by Diels and "saggezza'' 
by Walzer. In the variant version given by Stobaeus and repre- 
sented by our Fr. 80 it is intelligence {ro <j>povelv) that is described 
as common to all. 

Fr. 3 (D. 35; By 49)- XPl [y^p] ^^ y^Xa ttoAAwi/ tcrTopa<s 
<{>L\oa6c{>ov<i avS/oa? elvat. 

Clement of Alexandria Stromata 11. 421 (Stahlin). Cf. Din- 
dorf's ed., iii, p. 119. This is one of the earliest known uses of 
cf>i\6(7ocf>o'i, which is believed to have been of Pythagorean origin. 

Fr. 4 (D 22; By 8). xP^o-oi/ [yap] ol St^-^fievoi yrjv TToAA^v 
opvcraovo-L Kal evpicrKovatv oAtyov. 

Clement Stromata 11. 249. 

Fr. 5 (D 47> -^y 4^)' M ^'-'^V '"'^P'- "^^^ iivylaTtav crvfi^aWoijieda. 
Diogenes Laertius ix. y2. Quoted by Diogenes not in his essay 
on Heraclitus, but in the one on Pyrrho. 

[ 135 ] 

Fr. 6 (D 40; By 16). iroXvixaBir] voov cx^tv ov SiSacTKet. 

Diogenes Laertius ix. i (here, as elsewhere unless otherwise 
stated, a quotation from Diogenes is from his essay on Hera- 
clitus) and Clement, Stromata 11. 59. Aulus Gellius, Nodes At- 
ticae, Praefatio, 12, makes nearly the same quotation in Latin. 
Diogenes alone adds the phrase, accepted by Diels-Kranz, "other- 
wise it would have taught understanding to Hesiod and Pythag- 
oras, to Xenophanes and Hecataeus." 

Fr. 7 (D 108; By 18). 6k6(T(i>v Aoyovs ^/covo-a, ovSeU a^iKvetrat 

€9 TOVTO, a)(7T€ ytVOXTKetV OTL (T0(f>6v €(TTL TrdvTiDV Ke)((DpL(TfJieV0V. 

Stobaeus Florilegium iii. i. Here A070? in the plural evidently 
means "discourses." 

Fr. 8 (D lOI J By 80). iSi^rja-dfxrjv iixewvTOv. 

Plutarch Against Colotes 1118 C. 

Fr. 9 (D 116; By 106). dvOpMTroLcn iraai /xereaTt yivoiCTKeLV 
ko>vTOV<s Kal ao)cl>poveiv. 

Stobaeus Florilegium iii. 5. Diels alters the final word of 
the manuscript reading to cl>poveLv, and translates "klug zu sein." 
Diels-Kranz, followed by Walzer, restores the manuscript read- 
ing, as here. But Walzer, without stating why, questions the 
genuineness of the Fragment. 

Fr. 10 (D 112; By 107). o-(o</>/oomv dperrj iieytaTrj^ Kal (TO<f>Lr} 
dXr)Oea Aeyeiv Kal iroidv Kara (f>v(nv iwa'tovTa^. 

Stobaeus Florilegium iii. 5. Here again, although the manu- 
scripts read o-ox^/oomi/, Diels substitutes 4>povdv, while Diels- 
Kranz restores the manuscript reading. In this instance the dif- 
ference of interpretation is not so great, since Diels translates 
"das Denken" and Diels-Kranz "gesunde Denken." Walzer sides 
with Diels. In the case of both Frs. 9 and 10 I am unable to see 
sufficient ground for Diels' alteration of the manuscripts. Fair- 
banks believes that the two Fragments are not genuine. 

Fr. II (D 55' -^y ^3)" oo-toi/ 6\pi<i dKor) ixd6r)(n<s, ravTa eyw 


Hippolytus Refutation of All Heresies ix. Cf . Snell, in Hermes, 
61 (1926), p. 362. 

Fr. 12 (D lOI a; By 15)- OipOaX/xol [yap] twv wtwv dKpLJSeaTepoL 

Polybius XII. 27. 

Fr. 13 (D 107; By 4). KaKol ixdprvpe's dvOpMiroLcriv o^OaXixol Kal 
cDTa f3ap/3dpov<s xj/vx^'i i^ovToyv. 

Sextus Empiricus, Adversus Mathematicos vii ("Against the 

[ 136 ] 


Logicians," Bk. i), Sec. 126. Schleiermacher (fr. 22) translates 
the last three words as **mit rohen Seelen." 

Fr. 14 (D 73 J By 94). ov Set wa-n-ep Ka6ev8ovTa<; TroLelv kol kiyav. 

Marcus Aurelius iv. 46, where it follows the quotation of Fr. 


Fr. 15 (D 89; By 95). rois lyp-qyopoaiv Iva KaX kolvov koctixov 
elvaiy Twv 8e KOLfxwfievoiv cKaarov ct? lSlov d7rocrTp€^€(7^at. 

Plutarch On Superstition 166 C. 

Fr. 16 (D 21; By 64). Odvaros icTTiV OKoaa eyepOevre^ opiofxev, 
OKoaa 8c €v8ovT€<i V7rvo<;. 

Clement Stromata 11. 205. More literal translations are those 
of Freeman ("All that we see while slumbering is sleep"), Diels 
("Tod ist alles, was wir im Wachen sehen, und Schlaf, was im 
Schlummer"), and Walzer ("Tutto cio che vediamo dormienti e 
sonno"). But in taking care to reproduce the form of the Greek 
tautology these writers have lost the connotative force of wi/o?. 
Not only sleep is meant, but also the kind of awareness that may 
take place in sleep — i.e., dream. Thus in Plato {Laws vii. 800 A) 
we read: "if one of them had divined it vaguely KaO' vttvov' — 
which could be translated "in sleep," but equally well "in a 

Fr. 17 (D 123; By 10). cf>v(n<5 KprnrredOai (f>LXet 

Found in Themistius, Proclus, and twice in Philo. 
Fr. 18 (D 935 By 11). 6 ava$ ov to fxavrdov Icttl to Iv A€A<^ot? 
ovTe Aeyet ovtc KpvirTU dAAa arjiiaiveL. 

Plutarch On the Pythian Oracles 404 E. lamblichus De 
Mysteriis iii. 15, confirms the quotation, putting it in indirect 
discourse and using participles in place of verbs. The word 
(TTjfjiaLveL, which I have translated "gives signs," appears to have 
been derived from o-7)/xa, which had meant "a sign." In early 
Greek medical vocabulary the o-ij/xa was a symptom, and (r-qp^alveLv 
carried the idea of a symptom indicating or suggesting something 
more than itself. Later, as in Plato's Phaedo, the word (rrjfxa 
meant "tomb." This meaning appears to have developed out of 
the practice of placing a mark on the grave to indicate to the god 
who was buried there. The overtone of religious mystery may still 
have been present in Heraclitus' use of o-r^/xaiVetv, although the 
direction of communication is now from god to man, not from 
man to god. 

Fr . 19 ( D 18; By 7 ) • ^^.v fxrj eATnyrat dve Attiottov ovk i^evpT^creLy 
aveiepevvrjTov iov kol arropov. 

Clement Stromata 11. 121. kav firj ekirrjTai aviXinffTov OVK 

[ 137 ] 


iievprjaei . . . Fairbanks and Kathleen Freeman are wrong, I 
think, in taking this to mean, "If one does not hope, one will not 
find the unhoped for" (Freeman's translation). It does not sound 
like Heraclitus, and it is not a necessary translation. For eXTr-qrax 
need not mean ''hope" ; it can also mean "expect," and even "ex- 
pect anxiously" (Liddell and Scott). Heraclitus probably meant 
to steer his meaning between the extremes of hope and anxiety. 
The word aviXinGTov could be the object of either of the two 
verbs, or perhaps could belong rather loosely to both at once; 
I am taking it as primarily the object of eXir-qTai and this leaves 
the object of iievprjo-ei somewhat vague. I have supplied the word 
"truth" in order to complete the English sentence ; if taken loosely 
it cannot be too far from Heraclitus' meaning. 


Fr. 20 (D — J By — ). Trdvra pet, ovSev Se fxeveL' iravra ^(oypeL, kol 
ovSev fievei. 

The text as here reconstructed is drawn from several well- 
known passages, varying in expression but alike in purport. The 
second clause appears In the Cratylus, where Plato attributes it 
to Heraclitus with the words Aeyet on. In the same Dialogue, 440 
C, he attributes to the followers of Heraclitus (ot Trept 'Hpa/cXetrov) 
the phrase "Everything flows" {iravra pel), separating the two 
words by what is obviously his own disparaging comment, "like 
leaky pots." In the Theaetetus, 182 C, the word pel is coupled 
with KLvelrai to describe the Heraclitean doctrine, although 
Heraclitus is not mentioned here. Elsewhere in the Theaetetus 
(182 C-D and 160 D) Plato uses various cognate forms of the 
verb pelv to describe the Heraclitean doctrine; at 160 D he 
curiously ascribes the doctrine to both Heraclitus and Homer. 
Again, Aristotle in the Metaphysics (xiii 11078 b, 14-15) speaks 
of the doctrine that "all perceptible things flow" (w? iravTaiv twv 
alaOrjTiov ael peovroiv) and labels the doctrine "Heraclitean." Al- 
though such cumulative evidence may fall short of establishing 
the words as a direct quotation from Heraclitus, there is hardly 
any phrase that has been more widely associated with his doc- 
trine in classical times and by later doxographers ; and it is in- 
disputably Heraclitean in tone and substance. 

Fr. 21 (D 91, 12; By 41). Trora^w [yap] ovK ecTTLV eix/^rjvab 
81? Tw avTtp' erepa Kal erepa vhara hrippel. 

Plutarch, On the E at Delphi, 392 C. Variant versions are 

[ 138 ] 


given by Aristotle in the Metaphysics (iv: loioa, 13), by Plu- 
tarch in his Quaestiones Naturales (912 A), and by Arius Didy- 
mus (Dox., p. 471). The last author joins the Fragment some- 
what oddly with Fr. 44. Aristotle mentions the remark as if it 
were already well known as coming from Heraclitus. Diels uses 
Arius Didymus' version (from which the last five words are 
taken) as his fr. 12, Plutarch's as his fr. 91. 

Fr. 22 (D 126; By 39)* '^^ ^^XP^ diperat, Oepfwv if/vx^Tai, vypbv 
avatverai, Kap<f>a\€ov voTL^eraL. 

The Fragment is quoted by a scribe named Tzetzes in a 
commentary on the Iliad. I have not been able to verify it there, 
but the accuracy of the quotation is attested by Bywater, Diels- 
Kranz, Walzer, and others. 

Fr. 23 (D 84a; By 83). fxeTa/SdWov avaTraverat. 

Plotinus Enneads iv. viii. i. The Greek phrase might have 
either of two meanings. It might be a paradoxical way of saying 
that everything changes continually; or it might mean (cf. Fr. 
89) that there is comfort to be had in changing one's situation. 
My translation is guided by the former idea, and this is evidently 
justified by the preceding words in Plotinus: "Heraclitus . . . 
tells of compulsory alternation from contrary to contrary, speaks 
of ascent and descent, says that . . ." — and then comes the 
Fragment. On the other hand the Stephen McKenna translation 
of Plotinus, as revised by B. S. Page (London, 1953) renders 
the quotation, "Change reposes" ; which evidently implies the 
second interpretation. 

Fr. 24 (D 52; By 79)- <^^*^^ ttol'^ Ian Trai^Mv, Trecra-evoiV TratSo? ^7 

Hippolytus IX. The idea of the Trai? irait,<av, in varying forms, is 
attributed to Heraclitus by Clement, Proclus, and Lucian. 
Plutarch, without mentioning Heraclitus, speaks of "the poet's 
fancied child playing a game amid the sand that is heaped 
together and then scattered again by him" {On the E at Delphi 
393 E ; F. C. Babbitt's translation in the Loeb Classical Library 

The meaning of atw is in question. In Christian Greek vo- 
cabulary at the time of Hippolytus the word had come to mean 
"eternity." Some have held that in Heraclitus' usage it means 
"the great age" — i.e., the vast stretch of time between two world 
conflagrations; but others, and particularly those who assign a 
later Stoic origin to the theory of world conflagrations, reject so 
definite an interpretation. It seems best to follow the practice 
of Diels and Burnet in adopting the neutral translation, "time." 

[ 139 ] 


Diels-Kranz has altered Diels' ''Zeif to ''Lebensseit," and Fair- 
banks similarly writes "lifetime." 

Fr. 25 (D 53 5 By 44)- toAc/ao? iravTiov /xev Trar-qp €(tti, ttcivtwv 8e 
^acnXev^y kol tov<s fxkv 6eov<s eBet^e tov<; Se avOpwirov^, Tov<i fxev BovXovs 
eTTOLrja-e tov<s Se e\ev6epov<s. 

Hippolytus IX. In this and the following two passages the 
word "war" is used for ttoAc/xos and "strife" for epis. Kirk is 
evidently right in saying (p. 242) that the two words represent 
the same concept. 

Fr. 26 (D 80; By 62). elSevai Se xph "^^^ ttoAc/xov eovra iwov, Kal 
8lk7]v epiVy Kal yivofxeva Trdvra Kar epiv kol xP^'^v. 

Origen Against Celsus. The last phrase presumably means 
that the strife is a condition into which all things are driven by 
inner compulsion. Kirk (p. 242) offers the interpretation : "War- 
strife is everywhere, normal-course-of-events is war-strife, every- 
where things happen by war-strife and normal-course-of-events." 

Fr. 27 (D — ; By 43)- ^^^ '^^^ jaeja^crat Tw '^Ofji-qpoi ['HpaKAeiTO?] 
elTTOVTL ' w? epi<s eK t€ Oetov ek t avSpoiiroiv aTToXoiro ' otx'^o'eo-^ctt yap 
(fyrjCTL iravra. 

Simplicius, in his commentary on Aristotle's Categories : pub- 
lished in Commentaria in Aristotelem Graeca, Vol. viii, p. 412. 
Immediately before making the quotation Simplicius attributes 
to "the Heracliteans" the view that "if either of the opposites 
should fail, there would be complete and utter destruction of 

everything" : el yap to erepov T(j)v ivavTLOiv eTnXeixf/eL, olxolto av iravra 

acjiavLadevTa. According to Aristotle {Eudemian Ethics, vii: 
1235a, 26) Heraclitus supports his repudiation of Homer's re- 
mark by arguing that "there could be no harmony without both 
low and high notes, nor could life exist without both male and 
female." The Homeric reference is to Iliad xviii. 107. 

Because of Simplicius' indirect way of introducing the quota- 
tion ("Heraclitus blames Homer for saying ..."), a slight gram- 
matical reconstruction has been necessary in the translation. 


Fr. 28 (D 90; By 22). Trvpo? re avTajJioifir] ra Trdvra Kal irvp 
aTrdvToiv oKOicnrep xp^o-o^ xprjixara Kai xp^yj^-ctTwi/ ;j(pv(ros. 

Plutarch On the E at Delphi, 388 E. Diels' rendering suggests 
a somewhat similar interpretation to the one here offered: 
"Umsatz findet wechselweise statt des Alls gegen das Feuer und 
des Feuers gegen das All. . . ." Kirk (p. 345) interprets some- 
what differently : "All things are an equal exchange for fire and 

[ 140 ] 


fire for all things. . . ." He justifies this by the statement that ''avr- 
reinforces the idea of exact reciprocity in afiot^rj" and he inter- 
prets the simile as emphasizing the equality of the exchanges. I 
would suggest that there might be a secondary connotation of 
moral repayment, or requital, in di/ra/xoi/??}, as there is when 
Aeschylus uses forms of the verb avTaixcL^eaOaL. 

Fr. 29 (D 30; By 20). Koa-fxov rovBe, Tov avrov airavTiov, ovre Tt? 
6eu>v ovre avOpoiirwv eiroi-qGeVy dAA* rjv det kol ccttlv /cat ecnai, irvp 
aet^wov, aTrTOfxevov fierpa Kat aTro^rfievvvixevov jxerpa. 

Clement Stromata 11. 396. The two participles in the last 
clause should be taken in the middle voice, not in the passive; 
for Heraclitus does not think of the kindling and extinguishing 
as performed by any agency outside the fire itself. 

There is a general ambiguity in the Fragment. Does it mean 
that the universe as a whole becomes fiery and becomes ex- 
tinguished at different times? At first sight the interpretation 
appears plausible, since the subject-accusative is kocjiiov T6vh€. 
On this interpretation the word ixirpa would refer to regular 
periods of time, and Heraclitus would evidently be speaking of 
world-cycles. But since the question of whether he believed in 
the Stoic doctrine of world-cycles is a controversial one (dis- 
cussed in Chapter iii), we ought to translate so as to avoid 
prejudging the controversy if possible. For it might be that /xeVpa 
should be interpreted not temporally but quantitatively. That 
is, the word might refer to the degree to which the processes of 
kindling and extinguishing are constantly going on in the 
universe, while the universe itself remains an ever-living fire. 
Some such interpretation is made by those who wish to avoid 
ascribing the doctrine of world-cycles to Heraclitus. Such an 
interpretation would be consistent with the range of meaning of 
liirpa and it might seem to be supported by the phrase Trvp dei^wov, 
but I am unable to see how it can be consistent with the state- 
ment that the cosmos itself becomes kindled and extinguished. 
However, see Kirk's careful discussion of the probable meaning 
of KoV/xo? (pp. 314-324). 

Fr. 30 (D 65; By 24). [icaAet Se avTO^ XPV^H'^^^W '^^^ Kopov. 

Hippolytus IX. Both Diels and Bywater limit the quotation to 
the words "craving and satiety." Nevertheless it is perfectly plain 
from Hippolytus' context that avTo refers to fire, and the meaning 
of the phrase is indeterminable unless the subject is specified. 
For in his Refutatio the present quotation is immediately preceded 
by the sentence, Aeyct Be koI cf>p6vLfJLOV tovto elvat t6 mrvp Koi tt}? 


[ 141 ] 


Fr. 31 (D 91 ; By 40)- crKihvqcn Kal irdXiv (jvvayu, koX irpoaucn Kal 

Plutarch On the E at Delphi 392 C. In Plutarch this quotation 
follows the first clause of our Fr. 21, on the impossibility of 
stepping twice into the same river. Diels combines the two 
Fragments into his Fragment 91 ; but since they are apparently 
disparate in sense, and since Plutarch separates them by a few 
words of his own, I am presenting them here separately as Frs. 
21 and 31. However, Plutarch evidently sees, or thinks he sees, 
a connection between them. After quoting Fr. 21 he continues: 
"Nor is it possible to lay hold twice of any mortal substance in 
a permanent state ; by the suddenness and swiftness of the change 
in it there 'comes dispersion and, at another time, a gathering 
together'; or, rather, not at another time nor later, but at the 
same instant it both settles ino its place and forsakes its place; 
*it is coming and going.' " (Translated by F. C. Babbitt, in his 
Loeb Classical Library edition of Plutarch, v, p. 241.) The two 
verbs in the first half of the Fragment are active and transitive. 

Fr. 32 (D 31 ; By 21). wpb's Tpoiral irpStTOV ddXacraa, 6a\da(T7]<; 
Be TO jLiei/ rjixicrv yrj, to 8e rjpiKJv irprjcrT-^p. 

Clement Stromata 11. 396. 

■^^' 33 (-D 31 » -^y 23)' OdXaacra hia^ierai, Kal fxeTplerai €i<i rbv 
avTOv \6yov 6kolo<s irpoaOev rjv 17 yeveaOau yrj. 

Clement, loc.cit. Diels treats this as a continuation of Fr. 32. 
However, Clement quotes the two passages separately, with 
remarks of his own between, and the general similarity of subject 
matter does not prove that they stood together in Heraclitus' 

Fr. 34 (D y6] By 25). ^rj irvp tov yri<i Odvarov Kal drjp ^rj rbv 
7rvpo<; Odvarov, v8o)p t,rj tov depots OdvaTov, yrj tov i^Saros. 

Maximus of Tyre, xli, 4. Diels-Kranz retains only the first 
five and the last three words — evidently because of a supposed 
inconsistency with Fr. 32. See my discussion of this point in 
Chapter iii. Cf. Plutarch, On the E at Delphi, 392 C. 

Fr. 35 (D 64; By 28). to, Sc irdvTa olaKL^et K€pavv6<;. 

Hippolytus IX. In Fr. 120 Kv^epvdrei replaces the present verb, 
evidently expressing the same idea. 

Fr. 36 (D6; By 32). 6 '^Ato? veo5 e<^' vi^^PXI ^o-tiv. 

Aristotle Meteorologica 355 a, 14; confirmed by several other 
writers. A late writer Galenus {Dox., p. 626) further explains 
that Heraclitus regarded the sun as a burning mass (ai/e/x/xa), 

[ 142 ] 



whose rising is a process of kindling (eiaxpcs) and whose setting 
is a process of quenching (a-fSems) . 

Fr. 37 (D 3; By — ). [Trepl ix€ye$ov<s tjXlov] €V/oo? ttoSos 

Aetius, in Dox., p. 351. 

Fr. 38 (D 99; By 31). el fxrj ^Ato? -qv, eveKa rwv aAAwv aarpoiv 
€ucf>pov7) av rjv. 

Plutarch quotes this in his essay, **Is Fire or Water the More 
Useful?" (957 A). The Greek idiom is a little odd: eveKa r<ov 
a\\(i)v a(TTp(jDv evcfipovr) av rjv does not mean **on account of the 
other stars it would be night"; for this makes no sense. The 
force of eveKa is evidently weaker, and the indicated relation is 
about midway between the causal and the concessive; the idea 
being, "for all that the other stars could do. . . ." The word 
€v<f>p6vT] is a euphemism for "night." 

F^- 39 ("D 120; By 30). r]ov<s Kal ecnrepa<i Tepjiara rj apKTO'i Kal 
avTLOv rrfi apKTOV ovpo<i aWpiov Atos. 

Strabo i. i. 6. 

Fr. 40 (D 124; By — ). wo-Trep aappn eiKYJ Ke^vixevoiv 6 icaAAio-TOs 
[6] KocrpLO^. 

Theophrastus Metaphysics 15. 

Fr. 41 (D II ; By 55)- "^^^ \.y^p\ epireTOv TrAryy^ vefierai. 

Pseudo-Aristotle De Mundo 401a, 11. The Aristotelian writer's 
preceding clause (/cat cf>6eLpeTai toU tov Oeov TreiOofxeva OecrfjioU) indi- 
cates that he understands the quotation as referring to a divine 


Fr. 42 (D 45; By yi). if/vx^^ irelpaTa iwv ovK av e^evpoio, iraaav 
£7ri7ro/oeuo/xevo'? bhov' ovrm ^aOvv Xoyov eyei. 

Diogenes Laertius ix. 7. In De Anima, 11, Tertullian offers 
a somewhat shorter Latin version: "Terminos anintae nequa- 
quam invenies omnem viam ingrediens." 

Fr. 43 (D — ; By — ). [Kal 'H/oaKActro? 8e TYjv oLpyrjv elvai (f>ri(n] 
xf/vxTjy [etvep] Tr]v avaOvfJLLaacv, e| 17? raAAa crvviaTTjaiv. Kal acrw/xaTtura- 
Tov re Kai peov act* to 8e Kivovpievov KLvovfxevw yivo^a^KecrOaL. 

Aristotle De Anima 1 : 405 a, 25-28. In the introductory clause 
Aristotle is probably giving his own philosophical inference as 
to what Heraclitus must have meant by his statement that soul 
is a fiery vaporization out of which everything else is derived. 
R. D. Hicks in his translation of De Anima renders avaOvfiiaa-i's 
simply as "vapor" ; J. A. Smith in the Oxford translation renders 

[ 143 ] 


it as "warm exhalation." The latter rendering has the double 
advantage of connoting process and of suggesting some affinity 
between this process and fire. I have chosen the phrase "fiery 
vaporization" as preserving both connotations and making the 
latter more explicit. The transformation from water to air (which 
the word "vaporization," applied to ancient Greek cosmology, 
would presumably denote) and the transformation from air to 
fire are conceived by Heraclitus as continuous aspects of the 
Upward Way — so continuous that (as observed in Chapter iii) 
he sometimes finds it unnecessary to mention air as the inter- 
mediate state. 

Neither Diels nor Bywater accepts the Fragment as a direct 
quotation from Heraclitus, and it must be admitted that several 
of the words have an Aristotelian ring. But while Aristotle 
may have employed a little freedom of language, he is seriously 
reporting what Heraclitus taught; and the contained ideas are 
too important to ignore. 

Fr. 44 (D 12; By 41, 42)' [/cai] xjjvyal 8e airh T(ov vypdv avaOv- 


Arius Didymus (Dox., p. 471), where the sentence follows 
immediately after Fr. 21. 

Fr. 45 (D 115 5 By — ). i/'vx^? ecm Aoyo? eavTOv av^wv. 
Stobaeus Florilegium iii. i. 

Fr. 46 (D 118; By 75)* ^^V ^^Xl o-ocfxaraTr) Kal apL(TTr]. 

Stobaeus Florilegium iii. 5, as given by Stephanus, followed 
by Walzer and others. Diels-Kranz accepts the reading, avyr] 
irjpr] if/vxrj. . . , which could be translated, "A soul which is a 
dry beam of light is wisest and best." But although the light 
imagery is quite in keeping with Heraclitus' style of thought, 
Burnet argues cogently (p. 138, n. 2) that the avyrj irjprj is a 

Fr. 47 (D yy; By 72). xj/vxijcn ripif/LV [rj OdvaTov] vypyjai 

Numenius, by way of Porphyrius {The Cave of the Nymphs 
x). Here I follow By water's and Walzer's texts in omitting the 
two words as a probable interpolation. Diels-Kranz retains them, 
which would make the meaning "pleasure, which is to say death." 
It could be Heraclitus' own statement, but it sounds more like 
a copyist's attempt to make the implication explicit. In Por- 
phyrius, and evidently in the writer from whom he quotes, 
the Fragment is followed by the statement, "They live in each 

[ 144 ] 


Other's death and die in each other's life," which in the present 
canon serves (on the authority of Hippolytus) as the second 
half of Fr. 66. 

Fr. 48 (D 1 17 5 By 73)- o.vr]p okotov fx^BvaO^, ayerai viro TratSo? 
avq^ov arcf>a\\6fjL€VO<s, ovk ejratwv okt} jSaiva, vyprjv tt/v xpv^v ex*^^' 

Stobaeus Florilegium iii. 5. 

Fr. 49 (D. 36; By 68). if/vxfjcnv ddvaro^ vSiop yeveaOaL, vSart 
8e OdvaTO<i yrjv yeviorOai, Ik yrj'i Se v8(op yiverai, ei vSaro? 8e if/vxi]' 

Clement Stromata 11. 435. The first half of the quotation re- 
ceives confirmation from Philo and Hippolytus. 

Fr. 50 (D 125; By 84). Kat 6 kvkcwv Stio-Tarat firj KLvovfxevo<s. 

Theophrastus De Vertigine, 9. The firj has been added by 
Bernays to the manuscript text, and is obviously needed. 

Fr. 51 (D 85J By 105). Ov/ito /xa^^o-^at ^a^C'^'Oi'* o yap av OeXyj, 
xj/vxrj'i oivdrai. 

Plutarch Lije of Coriolanus 224 C. 

Fr. 52 (D no; By 104). dj/^pwTrot? yivecrOai oKoaa 6e\ov(nv ovk 

Stobaeus Florilegium iii. i. 

Fr. 53 (D 95; By 108). dfjLaOtTjv [yap] a/xctvov Kpvtrreiv, epyov 
8e Iv ave(T€L kol Trap' olvov. 

Plutarch Quaestiones conviviales 644 A. 

Fr . 54 ( I^ ^7 » By 117)- ^^^^ dvOpom^o^ ein Travrl Aoyw hrrorjcrdai 

Plutarch On the Right Method of Hearing 41 A. 

Fr. 55 (-D 34J By 3). divveroL aKovaavTe's KOicfiol(nv eoLKacn' ^dri<s 
avTolaiv fiapTvpel Trapcovra? (XTrctvat. 

Clement Stromata 11. 404. 

Fr. 56 (D 46; By 132). tt/v re ol-qa-iv iepdv voaov [eXeye Kal 
TTjv opaaiv ij/evSecrOaL] . 

Diogenes Laertius ix. 7. According to Liddell and Scott otr/o-t? 
can mean either (or both) opinion or self-conceit; the word 
"bigotry" can perhaps convey the double connotation. 

Fr. 57 (-D 17 > By ^), ov ydp <f>poveov(TL Totavra ttoXXol, okoctol 
lyKvpevcriv, ovhe pLaOovre^ yLvd)crKov(nv, loiVTolai Se SoKeovcn. 

Clement Stromata 11. 117. 

Fr. 58 (D 7> By 37)- ^^ irdvTa rd ovra Kanvo^ yevoiTO, plve>i av 

Aristotle De Sensu 443 a, 22. I translate pti/e? (literally "nos- 
trils") as "smell," because obviously, since all things are now 
supposed to be smoke, the word must be intended metaphorically. 

Fr. 59 (D 98; By 38). al xjJvxaX oafiiovTm KaO' "Al8yjv. 

I 145 ] 


Plutarch On the Face of the Moon 943 D. The verb ocr/xwi/rat 
means simply "smell," but the sense of the statement appears 
to be that this is how the souls of the dead, reduced to the 
condition of smoke, would have to perceive — an idea already 
indicated in the preceding Fragment. 

Fr. 60 (D 96; By 85). vIkve^ \.y^p\ KOTrp'nav iK^X-qroTepoi. 

Plutarch Quaestiones conviviales 669 A; Strabo, xvi. iii. 26. 

Fr. 61 (D yS; By ^6). ^do<s [yap] avOp^ireiov fi€v ovK e^ei 
yi/w/Aa?, Oeiov Be ex^L. 

Origen Against Celsus vi. 12. 

Fr. 62 (D — ; By — ). [koI firjv pr)Tu><; 6 *HpaKAetTO? (f>r)aL] ro 
fXTj elvai XoyiKov rov avOpoiirov, fiovov B' Wap^etv cj>pevrjpe<i to Trepiexov. 

Sextus Empiricus Adversus Mathematicos viii. 286; in the 
Loeb Classical Library edition. Against the Logicians 11. 286 
(in Vol. II of Sextus' writings). Although Diels and By water 
omit the Fragment from their lists, Sextus says that Heraclitus 
"expressly" (pTyrw?) affirms it. 

Fr. 63 (D 86; By 116). [aAAa] t<ov fih OeUv m TroAAa [Ka6' 
*HpaKAetTOv] aTTLarTLTj SLa(f>vyydv€i pirj yLyvoxxKecrdaL. 

Plutarch Coriolanus 232 D. 

Fr . 64 ( D 72 ; By 93 ) • ^ fiaXiara StTyve/cws ofXiXovaL Aoy w, tovt(o 
BLa(f>epovTaL, [kol oU KaO' -qfxipav iyKvpovac, ravra avTOLs ^eva <j)aiveTai] . 

Marcus Aurelius iv. 46. I follow Burnet (p. 139, n. 3) in 
supposing that the bracketed words belong to Marcus Aurelius 
and not to Heraclitus. 

Fr. 65 (D 26; By yy). av9po)7ro<; iv €v<f>p6vr) (f>do<5 aTTTerai eavTut* 
OLTroSavoiv ctTroo-ySecr^ets oi/^eis ^tov Be airTerai Te$vewTO<i ev8o)v' dTroa-peaOeU 
oipet^ eypr]yopoi<s airreTai evBovTO'i. 

Clement Stromata 11. 310. Cf. Dindorf's ed., 11, p. 399. 

Of all the Fragments of Heraclitus there is none that has 
given rise to more textual disputes. Bywater, who appears to 
have gone on the principle of discarding every word that any 
scholar had disputed, gives a truncated version of the Fragment, 
which Fairbanks, following him, translates: "Man, like a light 
in the nighttime, is kindled and put out." Their interpretation 
loses the force of the complex pun on aTTTcrat, explained in Chap- 
ter v. The present version of the text is based partly on Diels- 
Kranz, partly on Dindorf's edition of Clement's Stromata; the 
two stops indicate what I take to be the main divisions of thought. 

[ 146 ] 


Fr. 66 (D 62; By 67). aOavaTOi OvrjToi, dvqrol aOavaroiy ^wvTes 
Tov iK€LV(j)v Odvarov, tov Se cKeivwv jSiov TeOvewres. 

Hippolytus IX. Clement of Alexandria gives a different ver- 
sion : avOpoiiroL Oeoi, deol avSpw-rroL (Paedagogus I. 236 ; cf . Dindorf, 
Vol. I, p. 326). Since ''immortals" and "mortals" commonly 
meant, from Homer down, gods and men respectively, Clement's 
paraphrase is doubtless justified ; however, the paradoxical form 
given by Hippolytus is more in keeping with Heraclitus' typical 
style. Moreover, supposing that the two Christian writers are 
referring to the same passage, it seems more likely that one of 
them should have undertaken to explain the meaning to his 
readers in literal terms than that the other should have taken 
a plain statement from Heraclitus and dressed it up as a paradox. 
Although no verb is specified in either version, the verb "be- 
come" (rather than "are") appears to be justified by the second 
half of the Fragment. Evidently Clement disagrees, since he 
adds the remark, Aoyo? yap avTo<i, "for their meaning is the same." 

Fr. 67 (D 27J By 122). avOpwirovs fievei airoOavovra'S acraa ovk 
eX.7rovTaL ovSe Sokcovctlv. 

Clement Stromata 11. 312. 

Fr. 68 (D 63 j By 123). [evOa 8* eom] eTravLcrracrOaL Kal (f)v\aKa's 
yiveaOat iyepri ^wvroiv Kal veKpC>v. 

Hippolytus IX. The reference of "they" is uncertain. 

Fr. 69 (D 119; By 121). rjOo^ av9pwir<a Saifxcov. 

Stobaeus Florilegium iv. 40. Cf. also Plutarch Moral Essays 
999 E. 

Fr. yo (D 25; By lOl). fxopoi [yap] fxe^ove^ fJbi^ova<s ixoLpa<s 

Clement Stromata 11. 271. A pun on the somewhat different 
connotations of {wpoi and fxoipat. 

Fr. 71 (D 28; By 118). [/cat jiivTOL Kal] AiKT] KaraA-qij/tTaL xf/evSwv 
TCKTOva^ Kal fJidpTvpa<s. 

Clement Stromata 11. 331. 

Fr. 72 (D 66; By 26). Trdyra [yap] TO TTvp eireXOov Kpcvet Kal 

Hippolytus IX. The last verb (future middle of KaraXafipdveLv) , 
translated "will overtake" in Fr. 71 and here, could carry also 
the connotations of "catch by surprise" and "legally condemn." 

Fr. y^ (D 16; By 27). to firj Svvov ttotc ttw? dv TL<S XddoL', 

Clement Paedagogia 1. 216. 

Fr. 74 (-D — ,' By — ). elvai yap Kal IvravOa Oeov<;. 

Aristotle Parts of Animals 1. 5 : 645 a, 17. The reference of 

[ 147 ] 


the adverb was to a place by the stove, where some visitors un- 
expectedly found Heraclitus warming himself. 

Fr. 75 ("D 5» -^y 126). KaX rot? dyaAjaao-t 8e rovrioKTiv ev^ovTai, 
oKolov d TL<s 86fxoL(TL XccfyrjvevoLTO, ov TL yiVMdKUiv Oeov<; ovB' rjpiaa's 
oiTtve? eicri. 

Aristarchus Theosophia 68 ; Origen Against Celsus vii. 62. In 
Aristarchus it follows Fr. yS; in Origen it stands alone. Diels- 
Kranz retains Aristarchus' coupling. 

Fr. y6 (D 14; By 124, 125). WKrtTrdAoisr, /xttyot?, ^a/c^ot?, Arjvai?, 
fivcTTaLs. Tot [yctp] vofiL^ofxeva Kar av6p(i)7rov<s fivaTi^pia dncpwo-ri 

Clement Protrepticus i. 16. 

Fr. 77 ( ■'-^ 1 5 » -^y 1 27 ) • ^^ Z*^ y°^P Atovro-ft) TTop.irqv Ittolovvto Kal 
vfjLveov acrfxa alSoLOL(nv, avaiSeo-Tara eipyacrr' av' MVTO<i Se 'AcSt/s Kat 
Aioi/Do-09, oreo) fxaivovTai Kal Xrjva'L^ovaLv. 

Clement Protrepticus i. 26. 

Fr. 78 (D 5 >* By 130). KaOaipovrai 8* dAAo) aifiaTL ixtaLvofxevoL 
olov €i TL<s €is nrjXbv ifJb^a<s ir-qXw aTrovt^oiTO. iialveaOai 8' dv Bokoltj, et rts 
avTOi/ avOpo)Tr(DV i7n<f)pd(TaLT0 ovto) iroteovra. 

Aristarchus Theosophia. See Note to Fr. 75. 

Fr. 79 (-D 925 -'^y 12). ^[/SvWa 8e /xatvojaevo) aroixarL dyeAao-ra 
Kat dKaAAtoTTio-Ta Kat afivpLcrra (fyOeyyofxevrj x'-^^^^ ^"^^^ e^iKvetrai Trj 
KJxovfj Sta Tov ^eov. 

Plutarch On f/i^ Pythian Oracles 397 A. 


Fr. 80 (D 113 >* By ^1). iwov ia-n iracn to <f>poveeLV. 

Stobaeus Florilegium iii. i. 

Fr. 81 (D 114; By ^1). ^v vw Aeyovras Icryvpit^eaOai XPV '^'P 
iw^ TravTioVf oKOicnrep vofxw ttoXl's, kol ttoXv i<TxvpoT€p(o<;. Tpe(])ovTai yap 
7rdvT€<i ol dvBpoiireLoi vd/xot viro evos tov delov* Kparei yap toctovtov 
OKoaov IBiXei Kat iiapKei iraaiv koI Tre/otytVerat. 

Stobaeus Florilegium iii. i. There is a play on words: 
ivv vw ("with rational awareness") and iww ("to that which 
is common"). The pun is obscured by Diels-Kranz's reading of 
the second word as vow. Here and in the last line I follow 

Fr. 82 (D 44; By 100). pidyeadai XPV ''"^^ S^/aov vnep tov vofiov 
oKOidirep TCt^eo?. 

Diogenes Laertius ix. 2. 

Fr. 83 (D 33; By no), vo^io^ Kal PovXrj irdBeaOai evos. 

[ 148 ] 

Clement Stromata ii. 404. 

Fr. 84 (D 49; By II3)« ^^^ liwl fxvpioi lav a/oio-TO? XI' 

Galenus. Cf. Symmachus via Walzer: ''Heraclitum . . . qui 
summam laudis arbitrabatur placere uni si esset optimus.'' 

Fr. 85 (D 29; By IIl). alpevvraL [yap] ev avrl ajravTiov ol 
apL(TTOL, icAeo? aivaov Bv^rdv' ol Se ttoAAoi KCKoprjvTaL oKoio-irep KTi^vea. 

Clement Stromata 11. 366. 

Fr. 86 (D 24; By 102). ap7]icf)dT0V<s Oeol Ti/xwcrt kcll avOpiairoi. 

Clement Stromata 11. 255. 

Fr. 87 (D 28; By 118). SoKcovra [yap] 6 BoKLixo)TaTO<s yivwcTKCt 

Clement Stromata 11. 331. I follow the manuscript reading 
of the last word, which seems to me to make better sense than 
Diels-Kranz's arbitrary emendation. 

Fr. 88 (D 43; By 103). v^piv XPI (y^^vvvvai /xaWov ^ TrvpKdiriv. 

Diogenes Laertius ix. 2. 

Fr. 89 (D 84 b; By ^2^. Ka/Aaro? Igtl toi? avToU fxoxOelv Kal 

Plotinus Enneads iv. 8, where it follows Fr. 23. 

Fr. 90 (D 97; By II5)« ^cwe? [yap] Kara^avt^ovaiv S)v av (Jltj 


Plutarch Whether Old Men Should Engage in Politics 787 C. 

Fr. 91 (D 104; By IIl). tU yap avrwv v6o<; rj cf>p'^Vy S-j^/xcov 
aoi8oL<n TreidovTaL Kal StSao-KciAo) \p€L(jiVTaL OfiiXo) ovk elS6Te<s otl "ot 
TToAAot KaKOL, oXiyoi he ayaOoL" 

Proclus, Commentary on Plato's Alcibiades, i. 

Fr. 92 (D 56; By — ). l^rjTrar-qvrai ol avOpoiiroL irpo<s rrjv yvoycnv 
Tiov <f>av€pu>v 7rapaTrXr](7iO)<s '0/x7^pw, 05 iyevero tu>v *EAA?^v(ov <Tocf>(i)Tepo<i 
iravTOiv. Ikclvov re yap TratSe? </>^etpa5 KaraKTCtvovTCS l^iraT-qaav €t- 
TTOi/re?' ocra eihofxev Kal iXd^ofxev, ravTa dTroXeLirofxev, oaa Se ovre 
etSoixev ovt' iXd^ofjiev, Tavra <f}epofiev. 

Hippolytus IX. The story is also told in the anonymous De 
Vita Homeri (falsely ascribed to Plutarch), Section 4. 

Fr. 93 (D 42; By 119)' t^v re "Ofiepov diiov €K Totv ayd)vu)v 
eK^aXXeaOaL Kal paTri^eaOaL Kal 'Ap^^iXoxov oixo'nii<i. 

Diogenes Laertius ix. i. 

Fr. 94 (D 106; By 120). ['HpaKAciro? liriiTXrj^ev] 'HacoSo) ra? 
pXv dyaOas Trotov/xeVw ret? 8k <f>avXa<; w? dyvoovvTt (f>vcnv ■^fiepa'i aTrdcra^ 
fiiav ovaav. 

Plutarch Camillus 138 A. Seneca translates the content of the 
latter clause : Unus dies par omni est (Epistles xii. 7) . 

Fr. 95 (D 121; By 114). d^LOV 'E</>eo-tois rjPrjSbv dirdyiaaOaL 

[ 149 ] 


rrafTL Kal tol<s avi]fioL<s ttjv ttoXlv KaraAiTretv, oirtve? *Ep/x.o8cojoov avBpa 
€0)VTtov ovqiCTTOV l^i^aXov (f)dvTe<s* rnxeoiv firjSe els 6v'qLaT0<i eano, el 
8e fly, aXXy re Kal fier aAAwj/. 

Strabo xiv. i. 25 ; Diogenes Laertius ix. 2. 

Fr. 96 (D 125a; By — ). jxr] lirikiiroL Vfxd<i ttAovto?, 'E^eVtot, iV 
€$e\ey^0L(T9e irovrjpevoixevoL. 

Tzetzes, in a commentary on Aristophanes, as quoted by Diels- 
Kranz and Walzer. 

Fr. 97 (D 20; By 86). yevojxevoL ^(o€tv eOeXovm ixopovs t ex^tv 
[fiaXXov Be dvaTraveaOai] kol TratSas KaraAetVovo-t fx6pov<i yeveaSai. 

Clement Stromata 11. 201. Following MuUach and Walzer I 
regard the bracketed words as probably Clement's gloss. 


Fr. 98 (D 8; By 46). to dvri^ow o-vficfyepov kol Ik rdv ScacfyepovToiv 
KaXXicrr-qv a/3/xoviav[, Kal iravra Kar epLV yLveadai]. 

Aristotle Nicomachean Ethics viii:ii55 b, i. On the transla- 
tion of dpiJLovLa(-rj) see the note to Fr. 116. 

Fr. 99 ("D III > By 104). vovao's vyieirjv eiroL-qa-ev r}hv Kal dyaOov, 
At/jto? Kopov, Ka/xaro? dvdirav(nv. 

Stobaeus Florilegium iii. i. 

Fr. 100 (D 23 J By 60). Sikt^? ovofia ovk dv ^Secrav el Tavra firj rjv. 

Clement Stromata 11. 252. 

Fr. 1 01 (D 61 ; By 52). OdXaa-aa vBoip KaOapo)Tarov Kal [xiapoyra- 
Tov, LxOvorb fxev ironpLOV Kal (yoiTrjpiov, dvOp(i>7roL<s 8e dvoTov Kal oXiBpiov, 
Hippolytus IX. 
Fr. 102 (D 9> By 51)' ovovs avpixar' dv eXeaOai fxdXXov ^ xpvaov. 

Aristotle Nicomachean Ethics x. v: 1176 a, 8. 
Fr. 103 (D 37; By 53). Sues caeno, cohortales aves pulvere 
vel cinere lavari. 

Preserved in Latin by Columella viii. 4. 

Fr. 104 (D 82, 83; By 99, 98). iriO-qKuiv 6 /caAAto-ro? alaxpcxs 
dvOpoiiroiv yev€t (rvfi^aXXeiv' dv6po)7ro)v 6 (TOcf>(ji)TaTO<s tt/oo? 6cov TriOrjKO'i 
(f>avelTat Kal cro(f>La Kal /caAAet Kal rots aAAoi? TrdaLV. 

Plato Hippias Major 289 A-B. I have combined and treated 
as a single Fragment what both Diels-Kranz and Bywater list 
as separate quotations. In Socrates' discourse the two clauses 
are briefly separated, but apparently only to suit the conventions 
of conversation. Diels-Kranz prints the Fragment as a para- 
phrase rather than as a quotation, but accepts it as representing 
an opinion of Heraclitus. After the first clause Socrates adds 

[ 150 ] 


the comparison : *^ . . and the fairest pot is ugly compared with 
any maiden" ; Lattimore ascribes this to Heraclitus, but most 
other scholars take it to be Socrates' addition. 

Fr. 105 (D 79 J -^y 97)- o-^P v?/7rto5 y]Kov(j^ irpo^ 8aLixovo<; oKwcnrep 
7rat5 TT/oos av8p6<i. 

Origen Against Celsus vi. 12. Diels-Kranz, Bywater, and most 
other scholars agree in taking Trpo? with accusative in Fr. 104 
to indicate comparison with, but in Fr. 105 with the genitive 
to indicate a shift in point of view. The sense of SaificDv is at once 
vague and yet particular. Patrick's translation, *'the Deity," is 
certainly wrong. Kathleen Freeman tries to steer between the 
universal and the particular by employing the word ''divinity" 
without an article; Diels-Kranz says "Gottheif ; Walzer, "il 

Fr. 106 (D 102; By 61). tw fxlv Oeip KaXa iravra Kal dya^ct kol 
StKata, avOpoiTTOi Sf a pXv aSiKa WeiA7^(/>acrtv a 8e SiKata. 

Porphyrins Quaestiones Homericae, on Bk. iv of The Iliad, 
Both the definite article and the sense of the sentence show that 
the reference here is not to just any supernatural power, as in 
Frs. 104 and 105, but to the perspective that is universal and 

Fr. 107 (D 58; By 58). ol [yow] larpoi, re/xvovrc?, KaiovTes, [Travrrj 
/3aaavL^ovTe^ KaKw? toi)? dppworTo wra?, ] hrairiovrai fxrjBev d^tot {jll(t66v 
Xafx^avuv [Trapct rdv appiiXTTOVVTdiV^ . 

Hippolytus IX. Although Diels-Kranz takes Hippolytus' next 
words, ravTa e/oya^d/xevot, as part of the quotation, they strike me 
as belonging rather to Hippolytus' ensuing commentary. 

Fr. 108 (D 60; By 69)- o86<s dvw koltio ixta Kcu COVT77. 

Hippolytus IX. Cf. Tertullian Adv. Marc. 11. 28 : ''Quid enim 
ait Heraclitus ille tenebrosus? eadem via sursum et deorsum." 

Fr. 109 (D 103; By yo). iwov [yap] apyrj Kal 7repa<s iirl kvkXov 
[TTcpK^epeta?] . 

Porphyrius Quaestiones Homericae: on Bk. xiii of The Iliad. 
Diels-Kranz and Burnet accept the final word as a part of 
the quotation. On the other hand, since there is no other known 
instance of the word occurring in early Greek, other scholars 
have concluded that it is Porphyrins' addition. Wilamowitz (in 
Hermes, 62, p. 276) goes so far as to argue that the words 
€7rt kvkXov should also be omitted — a procedure that had already 
been taken by Bywater in his text and by Fairbanks in his trans- 
lation. But since both of the last two nouns of the manuscript 
version are in the genitive, it is possible to remove one of them 

[ 151 ] 


alone. I have followed Gigon, p. lOO, and Kirk, p. 113, in the 
present reading. 

Fr. no (D 49^ > By 81). noTafioU rots avTol<s e/xj8atvo/xeV re Kal 
ovK efx/Saivofxev' [et/xev re Kal ovk et/xei^] . 

Quoted by the late Greek grammarian and allegorist named 
Heraclitus, who wrote a commentary on Homer. The bracketed 
words, which I take to be an addition by the doxographer, are 
accepted by Diels-Kranz and Burnet. On the other hand, Sene- 
ca's version of the Fragment {Epistles Iviii. 23) is simply: *7w 
idem flumen bis descendimus et non descendimus." The word 
"bis" repeats the idea of Fr. 21 and may be omitted, since it 
does not appear in the Greek source just mentioned. Seneca 
adds, as his own explanation presumably: "Manet enim idem 
fluminis nomen, aqua transmissa est J' This seems more plausible 
than the Greek doxographer's interpretation. 

Fr. Ill (D 59> -^y 50) • 7va</>€to) 686's €v$ela Kal crKoXirj [xla earl 
Kal rf avTiq. 

Hippolytus IX. He explains: "The circular movement of the 
instrument in the fuller's shop called 'the screw' is straight and 
curved, for it revolves up and circularly at the same time." Pos- 
sibly this is what Heraclitus had in mind; but it may also be 
that he was playing upon the double meaning of eu^ij?, which 
can connote both "straight" and "right." Walzer is able to pre- 
serve the double connotation in Italian: "il percorso retto et 

Fr. 112 (D 10; By 59)- o'vvd\f/L€<s 6ka koX ov^ oka, (TVjJicfjepofjLcvov 
8ia<j>ep6fxevov, avvaSov SiaSov [/cat] ck TravTcov ev Kal €$ evos nravra. 

Pseudo- Aristotle De Mundo 396 b, 20. 

Fr. 113 (D 88; By 78). ramo [r hi] ^rjv Kal TeOvrjKCx; Kal 
[to] iypr]yop6<s Kal KaOevSov Kal viov Kal yrjpaiov' raSe yap /AeraTrecr- 
ovTa eKelvd icm KaKelva irdXiv fieTairea^ovTa ravTa. 

Plutarch A Letter of Consolation to Apollonius 106 E. The 
translation of ixeTaireaovTa Tavra as "sudden unexpected reversal" 
seems to find some support in Kirk, who writes (p. 147) ; "That 
lierairiirreiv is habitually used with this sense of sudden complete 
reversal is indicated especially by phrases like /xereTreTrroiKct ra 
TT/aay/xara (Lysias 20. 14; cf. Thucydides VIII 68; Plato Epistle 7 
325 A), meaning *a revolution had occurred.' " 

In making the first bracketed deletion I follow Wilamowitz 
(in Hermes, 62, p. 276) ; the second is more widely agreed 

Fr. 114 (D 57; By 35). hihd(TKa\o<i he irXeiGTOiv 'B.aioSo's' 

[ 152 ] 


Tovrov liriaTavrai TrXela-Ta eiSevai, oarri^ rjfjLepTjv Kal €V(f>pov7jV ovk 
eyLvoiCTKev* €(Ttl yap ev. 

Hippolytus IX. Cf. Hesiod, Theogony, 124 and 748. 

Fr. 115 (D 48; By 66). tw ovv to^m ovofxa fiio<s, epyov Se ^avaro?. 

From a Greek anonymous volume on Etymology which sets 
ySto? as a heading. Whereas here the point of the word-play is 
left to the reader's imagination, in another version (in Eusta- 
thius' commentary on the Iliad) the pun between **bow" (^8105) 
and "life" (^810?) is made explicit. 


Fr. 116 (D 54 j -By 47)' oLp/JLOvlr} a(f>avr]<s cfxLvep7J<; icpctTTcov. 

Hippolytus IX stated twice. How should apixovi-q be translated ? 
Fairbanks, Freeman, and Lattimore translate it "harmony," but 
Burnet prefers "attunement," since the Greek word does not 
imply harmony in the presentday sense — i.e., simultaneous 
sounds, or chords. So much is true, but on the other hand two 
musicians whom I have consulted opposed the word "attune- 
ment" because it suggests to them the preparatory tuning up 
of the instruments. Kirk argues (p. 224) that dpfiovir), which 
had come from apfio^eiv ("to fit together"), probably did not 
yet have a musical significance in Heraclitus' day; accordingly 
he translates, "An unapparent connexion is stronger than an 
apparent." Nevertheless, it is clear that Heraclitus intends a 
cosmic, archetypal significance, and it appears to me that this 
is better suggested by our word "harmony" than by any English 
alternative. The same argument applies to Fr. 117. 

Fr. 117 (D 51 j By 45)- ®^ ivvtacnv okw? SLa(f)€p6fji€vov eoiVTw 
ofjLoXoyeei' TraAtvrpOTro? apjiovLr) oKMcnrep t6$ov kol \vpr}<;. 

Hippolytus IX. By water offers as a separate fragment (No. 
56 in his list) Plutarch's version of the statement (On a Tran- 
quil Mind 473 F — 474 A) where the name of Heraclitus is not 
mentioned. The important difference among these several ver- 
sions of the aphorism is that whereas Hippolytus records the 
word iraXivT POTTO'S ("bending back"), Plutarch gives in its place 
the word 7raAivTovo§ ("inverse harmony"), both in the essay 
just mentioned and in his essay On I sis and Osiris, 369 B. Of 
course there may possibly have been two separate passages in 
Heraclitus' treatise, each employing one of the words, as By- 
water evidently supposes. But in making a choice I think it 
best to follow Diels-Kranz and Fairbanks in accepting TraXivrpoTro^ 

[ 153 ] 


instead of Burnet, Walzer, and Kirk in accepting TraAiVrovo?. Psy- 
chologically it appears more likely that there might have been 
a shift from the first word to the second, stimulated by the musi- 
cal simile that follows, than in the opposite direction. Cf. Plato 
Symposium 187 A. 

Fr. 118 (D 505 By l). ovk Ifxov aAAo, rov Xoyov aKov(TavTa<i 
OfioXoyeiv (To<j)Ov larw ev iravra ctvai. 

Hippolytus IX. Burnet's translation, "not to me, but to my 
Word," does not strike me as making good sense. It obscures the 
intended contrast, which is between the personal and the supra- 
personal. Adolf Busse, however, argues in favor of Burnet's in- 
terpretation, in his essay, "Der Wortsinn von Logos bei Hera- 
klit," in Rheinisches Museum, 75 (1926), pp. 203^. Busse points 
out that it was customary with the Greeks, "as often with Plato," 
to distinguish between oneself and the word which one speaks. 
Accordingly he translates Aoyo? in this instance as Wahrheits- 
heweis, adding parenthetically "(die Stimme der Wahrheit)." 
But Busse agrees that in Frs. i and 2 the primary reference 
is to the cosmic aspect ; and there he renders Aoyo? as Weltgesetz. 

Bruno Snell, in Hermes, 61, p. 365, argues along somewhat 
similar lines, stressing the connotation of "meaning" (meinen) 
in the words Aoyo? and Xiyuv. "Logos," according to him, refers 
to the real meaning of Heraclitus' words, and thus may be dis- 
tinguished from the words themselves with their accidental and 
personal characteristics. It is as if one were to say : "Don't merely 
listen to me, and to the sounds I make; attend rather to the 
meaning which they are intended to express." 

As for the word 6/xoAoyeti/, Kirk offers the view (pp. 67-68) 
that it contains a deliberate word-play. He argues that just as 
Aoyo9 means something much more than "word," so ofxoXoyeLv 
means more than "say the same word" or "agree," although 
it carries this sense too. But it also carries the connotation of 
being in agreement with the Logos, of not opposing the Logos 
by refusing to listen to it. Added weight is doubtless given to 
Kirk's theory by the use of the word which Hippolytus makes 
directly after the quotation : /cat on tovto ovk laaa-iv Travre? ovSe 
ofioXoyeovGiv, £7ri/>te/x<^erat wSe ttw? . . . 

Fr. 119 (D 32; By 65). ev to aocfybv fiovvov, XiyeaOai ovk eOeXet 
Kal iOeXei Zrjvo^ ovofxa. 

Clement Stromata 11. 404. The comma is not found in the 
manuscript, but is evidently required by the sense. 

[ 154 ] 


Fr. I20 (D 41; By 19). ev [yap] to (T0<f>6v, eTrto-rao-^at yvu)firjv 
OTTT) Kv^epvdrai iravra hia navTiov. 

Diogenes Laertius ix. i. The verb KvfSepvarai can be taken as 
either middle or passive in voice — as meaning "steer themselves" 
or as "are steered." Probably the distinction was not yet ex- 
plicit in Heraclitus' mind and vocabulary. Walzer writes "si 
governa" ; but the middle voice in Italian has a more nearly im- 
personal connotation than in English. 

For the variants of oTrrj, See Walzer, p. 80. 

Fr. 121 (D 67 J By 3^)- ^ ^^^^ Vl^^PV ^vcfypovrj, ;(ei/xa)i/ Bepo^y 
-TroAcjUO? elprjvrj, K6po<s At/xo?, dAAoiovrat 8e OKwaTrep <^ ^ 

OTTOTav (TvixjjLLyfj 6v(i)fJLa(nv, ovo/xa^crat KaO' rjSovrjv cKacrTOV. 

Hippolytus IX. The triangular brackets indicate the probable 
place of the main noun, missing from the manuscript. Since it 
was evidently dropped out by a copyist, there is no clue as to 
its nature or length. Several interpreters, including Diels-Kranz, 
Burnet, and Lattimore, have supplied the word "fire," and 
Burnet accordingly offers the translation, ". . . just as fire, 
when it is mingled with spices, is named according to the savor 
of each." There is no textual support for the interpretation. 

Among the other conjectures that have been offered are: 
"wine" (Schuster), "olive oil" (Snell), and "air" (Zeller). 
There is no textual support for any of them, and in my text 
and translation I have employed triangular brackets in order 
to avoid prejudging the question. Actually I am most inclined 
to agree with Hermann Frankel's theory, developed in his arti- 
cle, "Heraclitus on God and the Phenomenal World" (see Ap- 
pendix D, Pt. 3) and outlined in Chapter viii — namely, that 
Heraclitus may have been drawing a comparison with the an- 
cient practice of manufacturing unguents by blending a pure 
oily base with a concentrated odoriferous extract. On this hy- 
pothesis, the missing word would mean something like "neutral 
base" (oily, or possibly waxen), and Ovayfia would not mean 
either "incense" or "spices" (as various translators have inter- 
preted it) but "scent" or "fragrance." Cf. Plato Timaeus 50 C: 
"A substance that receives all bodies (o-to/xara) must itself be 

Fr. 122 (D 94; By 29). ^Ato? [yap] ovx vnep/S-qaeTaL [xerpa' 
el 8e fx-q, *Epivve? /xtv Aikt^^ eTTiKovpoi e^evp-^aovaiv. 

Plutarch On Exile 604 A. The same idea with slight verbal 
differences is found in Plutarch's On I sis and Osiris 370 D ; here 

[ 155 ] 


the handmaids are characterized as "stern-eyed" (yo/oyaWras . . . 
€7rLKovpov<s) and the term "Erinyes" is omitted. 

Fr. 123 (D 100; By 34). wpa? [at] iravra cfyepovm. 

Plutarch Platonic Inquiries 1007 D. 

Fr. 124 (D 75 J •'^y 9^)- '^tt^ '^^^'^ Ka6€v8ovTa<i e/ayara? elvat kol 
arvvepyov; twv iv rw Koafiio yivofieviov. 

Marcus Aurelius vi. 42. 


(Numbers refer to the Fragments. Consult the Index for writers not 
here mentioned.) 

Aristotle and his school : 20, 27, 36, 

41, 43, 44, 58, 74, 98, 102, 112. 
Clement of Alexandria : 3, 4, 16, 19, 

21, 29, 32, 33, 49, 55, 57, 65, 67, 

70, 71, 73, 76, 77, 83, 85, 86, 87, 

97, 100, 119. 
Diels, Doxographi Graeci: 37, 44. 
Diogenes Laertius: 5, 6, 42, 56, 82, 

88, 93, 120. 
Hippolytus: 11, 24, 25, 30, 35, 66, 

68, 72, 92, loi, 107, 108, III, 114, 

116, 117, 118, 121. 
Marcus Aurelius: 14, 64, 124. 
Maximus of Tyre : 34. 

Origen: 26, 61, 75, 105. 

Plato: 20, 104. 

Plotinus: 23, 89. 

Plutarch: 8, 15, 18, 21, 28, 31, 38, 

51, 53, 54, 59, 60, 63, 79, 90, 94, 

113, 122, 123. 
Polybius: 12. 
Porphyrins : 47, 106, 109. 
Proclus: 17, 91. 
Sextus Empiricus: i, 2, 13, 62. 
Stobaeus: 7, 9, 10, 45, 46, 48, 52, 

69, 80, 81, 99. 
Strabo: 39, 60, 95. 
Theophrastus : 40, 50. 

[ 156 ] 


The Fragments that follow are included in the Diels-Kranz 
list, but are omitted from the chapters of the present volume, 
as being either too trivial or obscure or insufficiently authorized 
to be of use in reconstructing the philosophy of Heraclitus. 

Fr. 125 (D 4; By — ). // happiness consisted in bodily pleas- 
ures, we would describe cattle as happy when they are eating 
fodder. Not found earlier than Albertus Magnus, in Latin trans- 

Fr. 126 (D 13; By 54). To delight in mud. popp6p<a xat/o«v. 
Athenagoras quotes these two words from Heraclitus, with the 
implication that Heraclitus, like himself, means that it is some- 
thing to be avoided. But there is not enough basis here for a 
full quotation. Cf. Fr. 103. 

Fr. 127 (D 19; By 6). Not knowing how to listen or how to 
speak. The subject of the sentence in Heraclitus is not stated. 

Fr. 128 (D 40; By 16). Otherwise it would have taught 
Hesiod and Pythagoras, Xenophanes and Hectaeus. A continua- 
tion of Fr. 6, in Diogenes Laertius as well as in Diels-Kranz and 
Bywater. But since Fr. 6 contributes to the topic of Chapter i 
whereas the present clause does not, it has seemed best to sepa- 
rate them. Moreover, the clause is added only by Diogenes 
Laertius ; omitted by Clement of Alexandria and Aulus Gellius. 
See note to Fr. 6. 

Fr. 129 (D 38; By 33). The first to study astronomy. Accord- 
ing to Diogenes Laertius, Heraclitus said this about Thales. 

Fr. 130 (D 39; By 112). Bias of Pirene, son of Tutamas, 
is of far greater account than the rest. Diogenes Laertius. Noth- 
ing else is known about Bias. 

Fr. 131 (D 68; By 129). Remedies. The rites of the Mys- 
teries, although shameful, are described by Heraclitus as "rem- 
edies" (lamblichus De Mysteriis i. 11). No doubt an interesting 
doctrine of Heraclitus' psychology may lurk here, but since only 
one word is quoted there is nothing solid on which to build a 

Fr. 132 (D 69; By 128). lamblichus {De Mysteriis v. 15) 
says that Heraclitus distinguishes two kinds of sacrifices : those 
which are performed by men who have first purified themselves 

[ 157 ] 


and those performed in an ordinary way. But Diels-Kranz takes 
the words to be those of lamblichus, not of Heraclitus; while 
Burnet rejects the passage altogether. 

Fr. 133 (D 70; By — ). Children's toys. According to lam- 
blichus, De Anima, Heraclitus so characterizes men's conjectures. 

Fr. 134 (D 71 ; By — ). He who forgets where the road leads. 
Marcus Aurelius (Meditations iv. 46) says we should take note 
of such a man. He is evidently quoting Heraclitus' phrase from 
memory, and there is no way of knowing how Heraclitus himself 
employed it. 

Fr. 135 (D 74; By — ). Like children to their parents. From 
Marcus Aurelius' reference (op.cit.) it appears that Heraclitus 
employed the phrase negatively, in warning against an attitude 
of uncritical acceptance. 

Fr. 136 (D 81 ; By — ). Leader of those who wrangle. Hera- 
clitus may have applied the phrase to Pythagoras, or to some 
group in the Pythagorean school. 

Fr. 137 (D 105; By — ). Homer was an astrologer. 

Fr. 138 (D 122; By 9). Approximations. 

[ 158 ] 




The two standard lists of Fragments, to one or the other of 
which nearly all scholarly references have been made during 
the last half century, are those of Bywater and Diels, the latter 
having been later revised by Kranz. The full titles of these three 
volumes (Bywater, Diels, and Diels-Kranz) will be found in 
Section 2 of the Bibliography. The Fragments as understood 
and translated in the present volume follow Diels-Kranz except 
where otherwise noted in Appendix B. Walzer's collection of 
the Fragments in Greek and Italian is likewise listed in Section 
2 of the Bibliography; but this collection should also be men- 
tioned here because of its unusual value in supplying, in addi- 
tion to the text of the Fragments, the doxographical context 
in which it appears, i.e., the sentences preceding and following 
it in the ancient writers from whom a given Fragment is quoted. 
Thus both Diels-Kranz and Walzer, although strictly speaking 
they are secondary sources, have served me, as they have served 
many another worker in the field, virtually in the capacity of 
primary sources ; for they provide a number of ancient materials, 
which might have been overlooked in going to the ancient 
sources alone, as well as a few quotations from rarely found 
authors who would otherwise have been difficult of access. 

My references to Aristotle's De Anima are to the edition and 
translation (Greek and English) of R. D. Hicks (Cambridge 
University Press, 1907). For the Metaphysics I have used the 
W. D. Ross edition (Oxford, 1924). For other parts of Aris- 
totle I have employed the Loeb Classical Library edition, sup- 
plemented by the Oxford University Press translations under 
the general editorship of Sir David Ross. 

Plato, Hippocrates, Diogenes Laertius, Plutarch, and Sextus 
Empiricus are quoted or translated from the Loeb Classical Li- 
brary editions of their writings. Sometimes the translations of 
these editions have been employed, sometimes the passages have 
been translated anew. In the case of Plato, Cornford's transla- 
tion of and commentary on the Theaetetus and Sophist were also 
used: Francis M. Cornford, Plato's Theory of Knowledge (Hu- 
manities Press, 1935; Liberal Arts Press, 1957). 

[ 159 ] 


The quotations from Heraclitus made by Hippolytus (the 
most valuable group from a single doxographical source that 
we possess) are all to be found in Book ix of The Refutation oj 
All Heresies. The following editions of this work were used : 

Hippolytus, Philosophumena, sive Haeresitum Omnium Con- 
futatio. Greek text ed. by Patrice Cruice with Latin translation 
(Paris, i860). 

Hippolytus, Philosophumena, or Refutation of All Heresies. 
Eng. tr. by F. Legge (London, 1921). 

Clement of Alexandria Opera edited by Otto Stahlin, in Die 
griechischen christlichen Schriftsteller der ersten drei Jahrhun- 
derte (Leipzig, 1905-1909; 3 vols, in four). 

Quotations from the late Greek doxographers Aetius, Arius 
Didymus, and Galenus are taken from Hermann Diels, Dox- 
ographi Graeci (Berlin, 1879, 1929). 

Quotations from the Greek commentators on Aristotle — Alex- 
ander Aphrodisiensis, Asclepius, and Simplicius — have been 
taken from Volumes i, vi, and vii respectively of Commentaria 
in Aristotelem Graeca (Berlin, 1882-1907). 

Stobaeus : as cited by Diels and Walzer, op.cit. 


The first collection of Heraclitus' Fragments was made by 
Friedrich Schleiermacher in 181 7, and the text of each Frag- 
ment was accompanied by a German translation and an exposi- 
tory discussion. Although Schleiermacher's list has subsequently 
been amplified and superseded, it is of unique historical im- 
portance: all later lists have stemmed from it, directly or 
indirectly, and moreover it made Heraclitus available to nine- 
teenth century philosophers and poets in something like an in- 
tegral form. Goethe and Nietzsche were both deeply influenced 
by its contents. Schuster's list, published in German in 1873, 
added a number of important Fragments to those collected by 
Schleiermacher, and for three decades thereafter it was the list 
to which most German workers in the field were likely to ap- 
peal. The main collections of Fragments since that time, listed 
chronologically, are as follows. For the sake of easy reference 
the name of the editor or translator in each case precedes the 
title, regardless of sequence on the title-page. The asterisk (*) 
indicates works of primary importance. 

[ 160 ] 


Schleiermacher, Friedrich, Herakleitos der Dunkle von Ephe- 
sos: in his Werke, 3. Abteilung, 2. Band, pp. 1-146. 
*By water, Ingram: Heracliti Ephesi Reliquiae (Oxford, 1877). 
Greek text and notes. Bywater attempts a loosely topical 

Mullach, Friedrich Wilhelm: Fragmenta Graecorum Philoso- 
phorum (Paris, 1883-1888). Greek text and Latin trans- 
lation of the early Greek philosophers. 

Patrick, G. T. W. : The Fragments of the Work of Heraclitus 
of Ephesus on Nature (Baltimore, 1889). English transla- 
tion and introductory essay, the Greek text being supplied 
in an Appendix. 
*Burnet, John: Early Greek Philosophy (London, 1892; third 
edition, 1920). My references are to the 1952 reprint of 
the third edition. Chapter iii is devoted to Heraclitus; 
Burnet's translation of the Fragments is based upon By- 
water's arrangement, but with independent judgments as 
to authenticity and interpretation. 
^Fairbanks, Arthur: The First Philosophers of Greece (Lon- 
don, 1898). Pp. 23-63 are devoted to Heraclitus. Greek 
text and English translation of the Fragments, based on 
Bywater's arrangement. 

Diels, Hermann: Herakleitos von Ephesos (Berlin, 1901). 
Greek text, German translation; a forerunner of the Her- 
aclitus material in the volume that follows. 
*Diels, Hermann: Die Fragmente der Vorsokratiker (Berlin, 
1903 ; 3 volumes) . In the fourth edition of this work (1922), 
the last edition for which Diels himself was responsible 
before his death, pp. 67-113 of Volume i are devoted to 
Heraclitus. Diels' "A" list comprises ancient statements 
about Heraclitus and his philosophy; the "B" list consists 
of what Diels judges to be authentic quotations, given in 
Greek text and German translation. Diels arranges his 
Fragments, except for the first two, which according to 
Sextus Empiricus had stood at the head of Heraclitus' 
treatise, according to the alphabetically ordered names of 
the ancient authors who have preserved them by quoting 

Bordrero, Emilio: Eraclito (Turin, 1910). Italian translation 
with discussion. 

Stohr, Adolf: Heraklit (Leipzig, 1920). Greek text and Ger- 
man translation. 

[ 161 ] 


Burckhardt, Georg: Heraklit, seine Gestalt und sein Kilnden 
(Zurich, 1925). Greek text and German translation. Re- 
edited (Wiesbaden, 195 1) with the title, Heraklit, Urworte 
der Philosophie. 

Snell, Bruno : Heraklit, Fragmente (Munich, 1926). Greek text 
and German translation. 

Solovine, Maurice: Heraclite d'Ephese (Paris, 1931). French 

Jones, W. H. S. : "Heraclitus on the Universe/' in Volume iv 
of the Works of Hippocrates (Loeb Classical Library, 
1931). Greek text and EngHsh translation. 
*Diels, Hermann and Kranz, Walther: Diels, Die Fragmente 
der Vorsokratiker, revised by Walther Kranz (Berlin, 
1934). Published as the fifth edition of Diels' work; but 
while it retains Diels' numbering of the Fragments, it often 
alters his translations and sometimes his text in important 
respects. Abbreviated "Diels-Kranz." 

Lattimore, Richmond: on pp. 1 19-128 of Matthew T. McClure, 
The Early Philosophers of Greece (New York, 1935). The 
translation of the Fragments, following Bywater's arrange- 
ment, by Lattimore; the ensuing discussion by McClure. 
*Walzer, Richard : Eraclito; Raccolta dei frammenti e traduzione 
italiana (Florence, 1939). Greek text, Italian translation; 
together with Greek texts of passages from ancient authors 
supplying different versions of the Fragments. Follows 
Diels' arrangement. 

Gaos, Jose: H eraclito (Mexico City, 1939). Spanish transla- 

Mazzantini, C. : Eraclito (Turin, 1945). Greek text, Italian 

Freeman, Kathleen: Ancilla to the Pre-Socratic Philosophers 
(Oxford, 1948). English translation of Heraclitus' Frag- 
ments, pp. 24-34, following Diels. 

Battistini, Yves: Heraclite d'Ephese (Paris, 1948). French 

Kirk, G. S. : Heraclitus, the Cosmic Fragments (Cambridge, 
1954). Text and translation, in doxographical context, of 
45 of the Fragments. 

Quiring, Heinrich : Heraklit; Worte tonen durch Jahrtausende 
(Berlin, 1959). Greek text, German translation. 

[ 162 ] 



Aurobindo, Sri, Heraclitus (Calcutta, 1941). In English. Gen- 
erally catalogued in libraries under the name "Ghose," 
which however is not used in Sri Aurobindo's published 

Bernays, Jacob, Gesammelte Ahhandlungen, Vol. i (Berlin, 
1885). The first volume consists of articles on Heraclitus. 

Bernays, Jacob, Die heraklitischen Brief e (Berlin, 1869). Al- 
though the letters which Diogenes Laertius ascribes to Her- 
aclitus are admittedly spurious, Bernays approaches them 
guided by the question, "What kind of evidence can be 
obtained from spurious writings?" 

Bise, Pierre, La politique d'Heraclite d'Ephese (Paris, 1925). 

Brecht, J., Heraklit; ein Versuch iiber den Ursprung der Phil- 
osophie (Heidelberg, 1936). 

Cuppini, Noemi, Esposidone del sistema di Eraclito (Rome, 

Dauriac, Lionel, De H eraclito Ephesio (Paris, 1878). 

Frankian, Aram M., Heraclite. Appears as Vol. i of the author's 
Etudes de philosophie presocratique (Paris, 1933). 

Gigon, Olof, Untersuchungen zu Heraklit (Leipzig, 1935). 

Gladisch, August, Herakleitos und Zoroaster (Leipzig, 1859). 

Gomperz, Theodor, Zu Heraklits Lehre und den Ueherresten 
seines Werkes (Vienna, 1887). 

Herr, Alfred, Beitrdge Bur Exegese der Fragment e des Hera- 
kleitos von Ephesos (Eger, 1912). 
*Kirk, G. S., Heraclitus, the Cosmic Fragments (Cambridge, 
Eng., 1952). The most important book on Heraclitus that 
has appeared in English. 

Kirk, William C, Fire in the Cosmological Speculations of 
Heraclitus (Minneapolis, 1940; Princeton Ph.D. thesis). 

Lassalle, Ferdinand, Die Philosophie Herakleitos des Dunklen 
von Ephesus, 2 vols. (Berlin, 1858). Also published as 
Vols, vii-viii of his Gesammelte Reden und Schriften (1919- 

Macchioro, Vittorio, Eraclito; nuovi studi suU'orfismo (Bari, 

Mayer, G., Heraklit von Ephesus und Arthur Schopenhauer 
(Heidelberg, 1886). 

Mohr, J., H eraklitische Studien (Zweibrucken, 1886). 

Patin, Alois, Heraklits Einheitslehre (Leipzig, 1866). 

[ 163 ] 


Patin, Alois, Quellenstudien zu Heraklit; pseudohippokratische 
Schriften (Wiirzburg, 1881). 

Pfleiderer, Edmund, Die Philosophie des Heraklit von Ephesus 
im Lichte der Mysterienidee (Berlin, 1886). 

Pressley, G., Die metaphysischen Anschauungen Heraklits von 
Ephesus (Magdeburg, 1908). 

Rivier, Andre, Un emploi archdique de Vanalogie chez Hera- 
clite et Thucydide. (Lausanne, 1952). 

Schafer, G., Die Philosophie des Heraklit von Ephesus und 
die moderne Heraklitforschung (Leipzig, 1902). 

Schultz, Wolfgang, Pythagoras und Heraklit (Leipzig, 1905). 
Published as Vol. i of his Studien zur antiken Kultur 
(Leipzig, 1905-1907). 

Schuster, P., Heraklit von Ephesus (Leipzig, 1873). 

Sloninsky, H., Heraklit und Parmenides, published as Vol. vii, 
Hejt I, of Philosophische Arbeiten, edited by Hermann 
Cohen and Paul Natorp (Giessen, 1912). 

Spengler, Oswald, Der metaphysische Grundgedanke der her- 
aklitischen Philosophie (Halle, 1904). 

Spengler, Osvaldo, Herdclito (Buenos Aires, 1947). The first 
half of the volume consists of a long essay by Rodolfo 
Mondolfo discussing the views of eleven other Heraclitean 
scholars. There follows a Spanish translation by Augusta 
de Mondolfo of Spengler's aforementioned essay. 

Surig, Henrik Wilhelm, De Betekenis van Logos hij Hera- 
kleitos vol gens de Traditie en de Fragmenten (Nijmegen, 
1951). With a summary in English and an extensive bibli- 

Teichmiiller, Gustave, Herakleitos : Vol. i of his Neue Studien 
zur Geschichte der Begriffe (Gotha, 1876-1879). 

Weerts, Emil, Heraklit und Heraklit eer (Berlin, 1926). 


Auerbach, Walter, "Zur Gegeniiberstellung von Sein und 

Schein bei Heraklit/' Eos, 33 (1930), pp. 651-664. 

; "De principio heraclito," Eos, 32 (1929), pp. 301-314. 

Binswanger, Ludwig, "Heraklits Auffassung des Menschen," 

Die Antike, 11 (i935)> PP- i-39- 
Brieger, A., "Die Grundziige der heraklitischen Physik," 

Hermes, 39 (1904), pp. 182-223. 
Busse, Adolf, "Der Wortsinn von Logos bei Heraklit," Rhein- 

isches Museum, 75 (1926), pp. 203^. 

[ 164 ] 


Calogero, Guido, "Eraclito," Giornale critico di filosifia italiana, 

1936, pp. 195-224- 
Capelle, W., *'Das erste Fragment des Herakleitos," Hermes, 

59 (1924), pp. 190-203. 

Cataudella, Q., "L'armonia invisibile di Eraclito," Sophia, 17 

(1949), pp. 332-333' On Fr. 116. 
Cherniss, Harold, review of Olof Gigon, Untersuchungen zu 

Heraklit, American Journal of Philology, 56 (1935), pp. 

Deichgraber, Karl, "Bemerkungen zu Diogenes' Bericht iiber 

Heraklit," Philologus, 93 (1938), pp. 12-30. 
Diels, Hermann, "Heraclitus," Hastings' Encyclopaedia of Re- 
ligion and Ethics, 6, pp. 591-594. 
Disandro, C. A., ''Heraclito y el Lenguaje," Revue de Argue, 

. 3 (^954)- 
Fiore Sole, G., "II problema di Dio in Eraclito ed Eschilo," 

Sophia, 16 (1948), pp. 203-205, 357-361. 
Frankel, Hermann, "Heraclitus on God and the Phenomenal 

World," American Philological Association, Proceedings, 

69 (1938), pp. 230-244. 
— , "Heraclitus on the Notion of Generation," American 

Journal of Philology, 59 (1938), pp. 89-91. 
, "A Thought Pattern in Heraclitus," American Journal 

of Philology, 59 (1938), pp. 309-337- 

-, German version of his articles on Heraclitus, in his 

Dichtung und Philosophic des frUhen Griechentums (New 

York, 1938), pp. 474-505. 
Friedlander, Paul, "Herakliti Frag. 124," American Journal 

of Philology, 63 (1942), p. 336. This is our Fr. 40. 
Glasson, T. F., "Heraclitus' Alleged Logos Doctrine," Journal 

of Theological Studies, 3 (London, 1952), pp. 231-238. 
Gomperz, Heinrich, "Ueber die urspriingliche Reihefolge eini- 

ger Bruchstiicke Heraklits," Hermes, 58 (1923), pp. 20-56. 
Gregoire, F., "Heraclite et les cultes enthousiastes," Revue 

neoscholastique de philosophic, 38 (Louvain, 1935), pp. 

Heidegger, Martin, "Heraklit," 'AvTuSopov. Festschrift zur Feier 

des ^^ojahrigen Bestands des Heinrich-Suso-Gymnasiums 

in Konstanz (1954). 

, "Logos," Vortrdge und Aufsdtze, pp. 207-230. 

Hoelscher, U., "Der Logos bei Heraklit," Festschrift fUr Karl 

Reinhardt (Cologne, 1952). 

[ 165 ] 


Howald, E., "Heraklit und seine antiken Beurteiler," Neue 
Jahrbucher jiir die klassischen Altertumer, 27 (1918). 

Kerschensteiner, Jula, "Der Bericht des Theophrast iiber Hera- 
klit," Hermes, 83 (1955), pp. 385-411- 

Kirk, G. S., "Heraclitus and Death in Battle," American Jour- 
nal of Philology, 70 (1949)* PP- 384-393- 

, "Natural Change in Heraclitus," Mind, 60 (1951), 

PP- 35-42. 
Kranz, Walther, "Der Logos Heraklits und der Logos des 

Johannes," Rheinisches Museum, 93 (1949), pp. 81-95. 
Leuze, O., "Zu Heraklit Frag. 26 (Diels)," Hermes, 50 

(1915), pp. 604-625. This is our Fr. 65. 
Loew, Emmanuel, "Heraklit von Ephesus: die Entstehung 

des empirisch-physikalischen Wegens der Forschung/' 

Rheinisches Museum, 79 (1930), pp. 123-152. 
, "Heraklit im Kampfe gegen den Logos," Archiv zur 

Geschichte der Philosophic, 23 (1910), pp. 89-91. 
, "Das heraklitische Wirklichkeitsproblem und seine 

Umdeutung bei Sextus," Jahresbericht des Sophiengym- 
nasiums in Wien (1914). 

, "Das Lehrgedicht des Parmenides eine Kampfschrift 

gegen die Lehre Heraklits," Rheinisches Museum, 79 
(1930), pp. 209-214. 

"Parmenides und Heraklit im Wechselkampfe," Archiv 

zur Geschichte der Philosophic, 24 (1911), pp. 343-369. 

, "Das Verhaltnis von Lehre und Logik bei Heraklit," 

Wiener Studien, 51 (1933), pp. 14-30. 

"Die Zweiteilung in der Terminologie Heraklits," 

Archiv zur Geschichte der Philosophic, 24 (1911), pp. 1-21. 

Maddalena, A., "Eraclito nell'interpretazione di Plato e d'Aris- 
totele," Atti dell'Istituto Veneto di Scienze, Letter e ed 
Arti, 98 (Venice, 1938-1939), pp. 309-335- 

Merlan, Philip, "Ambiguity in Heraclitus," Actes du Xleme 
congres international de philosophic, 12 (1953), 56-60. 

, "Heraclitus fr. B93D," American Philological Asso- 
ciation, Transactions and Proceedings, 80 (1949), p. 429. 
This is our Fr. 18. 

Minar, F. L., "The Logos of Heraclitus," Classical Philology, 

34 (1939). PP- 323^- 
Mondolfo, Rodolfo, "Evidence of Plato and Aristotle relating 
to the ekpyrosis of Heraclitus," Phronesis, 3 (1958), pp. 

[ 166 ] 


Muth, R., "Herakleitos, I. Bericht: 1939 bis 1953," Anzeiger 
fur die Altertumswissenschajt, 7 (Vienna, 1954), pp. 65-90. 

, "Heraklits Tod," op.cit., y (i954), PP- 250-253. 

Nestle, W., ''Heraklit und die Orphiker," Philologus, 64 
(1905), pp. 367-384. 

, "War Heraklit Empiriker?" Archiv zur Geschichte 

der Philosophie, 25 (1912), pp. 275ff. 

Paci, E., "La concezione mitologico-filosofico del logos in 
Eraclito," Acme, 2 (1949), pp. 176-201. 

Pfleiderer, Edmund, "Die pseudoheraklitischen Briefe und ihr 
Verfasser," Sonder-Abdruck aus Rheinisches Museum fur 
Philologie, n.s., 42. 

Power, O. S., "Heraclitus, Fr. 28 Diels; a New Interpreta- 
tion," American Philological Association, Transactions and 
Proceedings, 78 (1947), pp. 432-433. These are our Frs. 87 
and 71. 

Rabinowitz, W. Gerson, and W. I. Matson, "Heraclitus as 
Cosmologist/' Review of Metaphysics, 10 (1956), pp. 244- 


Reinhardt, Karl, "Heraclitea," Hermes, yy (1942), pp. 225- 

, "Heraklits Lehre vom Feuer," Hermes, yy (1942), 

pp. 1-27. 

Robertson, D. S., "On the Story of Heraclitus told by Aris- 
totle in De Par tibus Animalium," Cdimhndge (Eng.) Philo- 
logical Society, 1938. 

Schultz, Wolfgang, "Die Kosmologie des Rauchopfers nach 
Fr. 67," Archiv zur Geschichte der Philosophie, 22 (1909), 
pp. 197-229. This is our Fr. 121. 

Snell, Bruno, "Die Sprache Heraklits," Hermes, 61 (1926), 

PP- 353-381. 
, "Heraklits Fragment 10," Hermes, 76 (1941), pp. 84- 

Sy. On Fr. 112. 
Spengler, Oswald, "Heraklit," in his Reden und Aufsdtze 

(Munich, 1951). 
Stefanini, L., "In nascita del logos in Eraclito," Giornale critico 

della filosofia italiana, 5 (1951), pp. 1-24. 
Tannery, Paul, "Heraclite et le concept de Logos," Revue 

philosophique, 16 (1883), pp. 292ff. 
Vernenius, W. J., "Psychological Statement of Heraclitus," 

Mnemosyne, Series iii (1943). 

[ 167 ] 


Vlastos, Gregory, "On Heraclitus," American Journal of Phi- 
lology, 76 (1955), pp. 337-368. 

Wolf, E., "Der Ursprung des abendlandischen Rechtsgedan- 
kens bei Anaximander und Heraklit," Symposium, i 
(1948), pp. 35-87. 

Wundt, M., **Die Philosophic des Heraklit von Ephesus im 
Zusammenhang mit der Kultur loniens," Archiv zur Ge- 
schichte der Philosophie, 20 (1907). 

Ziller, W., "Zu einigen Fragm. der heraklitischen Physik," 
Rheinisches Museum, 62 (1907), pp. 54-60. 

Zoumpos, A. N., *'Die metaphysische Bedeutung des Wortes 
*Ai8>;5 bei Herakleitos," Actes du XT erne congres interna- 
tional de philosophie, 12 (1953), pp. 34-55- 


Adam, James, The Religious Teachers of Greece (Edinburgh, 

, The Vitality of Platonism (Cambridge, Eng., 1911). 

Chapter iii: "The Doctrine of the Logos of Herachtus." 
All, Anthony, Geschichte der Logosidee (Leipzig, 1896). 
Anrich, Gustav, Das antike Mysterienwesen in seinen Einfluss 

auf das Christ enhild (Gottingen, 1894). 
Baeumker, Clemens, Das Problem der Materie in der grie- 

chischen Philosophie (Miinster, 1890). 
Bapp, Karl, Aus Goethes griechischen Gedankenwelt (Leipzig, 

Baric, G. T., Vesigenza unitaria da Talete a Platone (Milan, 


Beare, W., Greek Theories of Elementary Cognition (Oxford, 

Brandon, S. G. F., Time and Mankind; an historical and philo- 
sophical study of mankind's attitude to the phenomena of 
change (London, 1951). 

Buber, Martin, "What Is Common to All," Review of Meta- 
physics, II (1958), pp. 359-379- 

Burnet, John, Early Greek Philosophy (London, 1892; 4th 
edition, 1920). 

Calogero, Guido, Studi sulV eleatismo (Rome, 1932). 

Capelle, Wilhelm, Die griechische Philosophie, 3 vols. (Berlin, 
1922-1924) : Vol. I. 

[ 168 ] 


-, Die Vorsokratiker (Leipzig, 1935). 

Chauvet, Emmanuel, La philosophie des medians grecs (Paris, 

Cherniss, Harold, Aristotle's Criticism of Presocratic Philos- 
ophers (Baltimore, 1935). 

, "The Characteristics and Effects of Presocratic Philos- 
ophy," Journal of the History of Ideas, 12 (1951), pp. 319- 

Cornford, Francis M., From Religion to Philosophy (1912). 
, The Laws of Motion in Ancient Thought (Cambridge, 

Eng, 1931). 
, "Mystery Religions and Pre-Socratic Philosophy," 

The Cambridge Ancient History, 4 (1926), pp. 522-578. 
-, The Origins of Greek Philosophical Thought (Cam- 

bridge, Eng., 1952). 
Croissant, Jeanne, "Matiere et changement dans la physique 

ionienne," Antiquite classique, 13 (1944), pp. 61-94. 
Deichgraber, Karl, "Hymnische Elemente in der philosophi- 

schen Prosa der Vorsokratoker," Philologus, 88 (1933), 

PP- 347-361. 
Delatte, A., Les conceptions de V enthousiasme chez les philo- 

sophes presocratiques (Paris, 1934). 
Dies, A., Le cycle mystique: la divinite, origine et fin des exis- 
tences individuelles dans la philosophique antesocratique 

(Paris, 1909). 
Dodds, E. R., The Greeks and the Irrational (Berkeley, 1951). 
Eliade, Mircea, The Myth of the Eternal Return (Eng. tr., 

New York, 1954). 
Fairbanks, Arthur, The First Philosophers of Greece (New 

York, 1898). 
Freeman, Kathleen, The Pre-Socratic Philosophers (Oxford, 

Friedrich, Carl, Hippokratische Untersuchungen (Berlin, 


Garcia Bacca, Juan David, Los Presocrdticos (Mexico City, 

Gentile, M., La metafisica presofistica (Padua, 1939). 
Gigon, Olof, Der Ursprung der griechischen Philosophie von 

Hesiod bis Parmenides (Basel, 1945). 
Gilbert, Otto, Griechische Religions-Philosophic (Leipzig, 


[ 169 ] 


Gomperz, Theodor, Greek Thinkers (Eng. tr., 1901), Vol. i. 
Guerin, P., L'idee de justice dans la conception de I'univers 

chez les premiers philosophes grecs de Thales a Heraclite 

(Paris, 1934). 
Hack, Roy Kenneth, God in Greek Philosophy to the Time 

oj Socrates (Princeton, 1931). 
Heidegger, Martin, Einfiihrung in die Metaphysik (Tiibingen, 

1953) : esp. pp. 96-103, 127, 130, 146. 
, "Aletheia," Vortrdge und Aujsdtze (Pfullingen, 1954), 

pp. 257-282. On Fr. 73. 
Heidel, W. A., Hippocratica, i: Harvard Studies in Classical 

Philology, 25 (1914). 
, **On Certain Fragments of the Presocratics," Ameri- 
can Academy of Arts and Sciences, Proceedings, 48 (1913), 

pp. 681-734. 
, "On Physis; A Study of the Conception of Nature 

among the Pre-Socratics," American Academy of Arts and 
Sciences, Proceedings, 45 (1910), pp. 79-133. 

'Qualitative Change in Pre-Socratic Philosophy," 

Archiv zur Geschichte der Philosophic, 19 (1906), pp. 333- 


Heinze, Max, Die Lehre vom Logos in der griechischen Philos- 
ophic (Oldenburg, 1872). 

Hirzel, R., Themis, Dike und Verwandtes (Leipzig, 1907). 

Huit, Charles, La philosophic de la nature chez les ancients 
(Paris, 1901). 

Inge, William, "Logos," Hastings' Encyclopaedia of Religion 
and Ethics. 

Jaeger, Werner, Paideia (Eng. tr., New York, 1943), Vol. i. 

, Theology in the Early Greek Philosophers (Eng, tr., 

Cambridge, Mass., 1947). 

Joel, Karl, Geschichte der antiken Philosophic (Tiibingen, 
1921), Vol. I. 

, Der Ursprung der Natur philosophic aus dem Geiste 

der Mystik (Jena, 1906). 

Kafka, G., Die Vorsokratiker (Basel, 1948). 

Kirk, Geoffrey S., "The Problem of Cratylus," American Jour- 
nal of Philology, 72 (1951), pp. 225-252. 

, and J. E. Raven, The Presocratic Philosophers (Cam- 
bridge, Eng., 1957). 

Kranz, Walther, Kosmos und Mensch in der Vorstellung des 
friihen Griechentums (Gottingen, 1938). 

[ 170 ] 


Maddalena, Antonio, Sulla cosmologia ionica da Talete ad 
Eraclito (Padua, 1941). 

McClure, Matthew T. : see Lattimore in Section 2 of Bibli- 

Mondolfo, Rodolfo, Vinfinito nel pensiero dei greet (Florence, 

Mugler, Ch., Deux themes de la cosmologie grecque; devenir 

cy clique et pluralite des mondes (Paris, 1953). 

Navarro Monzo, Julio, La Biisqueda Presocrdtica (Monte- 
video, 1926). 

Nestle, Wilhelm, Vom Mythos zum Logos (Stuttgart, 1940). 

Nietzsche, Friedrich, *'Die Philosophic im tragischen Zeitalter 
der Griechen," Werke (1896), Vol. 10. 

Oehler, Richard, Nietzsches Verhdltnis zur vorsokratischen 
Philosophie (Halle, 1903). 

Onians, R. B., The Origins of European Thought about the 
Body, the Mind, the Soul, the World, Time and Fate (Cam- 
bridge, Eng., 1951). 

Palme, Adolf, Studien zur hippokratischen Schrift, irepl Siatrrys 
(Tiibingen, 1933). 

Patin, Alois, "Parmenides im Kampfe gegen Heraklit," Jahr- 
huch fUr die klassische Philosophie, 25 (1899), pp. 491-660. 

Reinhardt, Karl, Parmenides und die Geschichte der griechis- 
chen Philosophie (Bonn, 191 6). 

Rey, Abel, La science dans Vantiquite (Paris, 1930- 1948), 
Vol. II. 

Rivaud, Albert, La principe du devenir et la notion de la ma- 
tiere dans la philosophie grecque depuis les origines jusqu'd 
Theophraste (Paris, 1906). 

Rohde, Erwin, Psyche; the Cult of Souls and Belief in Immor- 
tality among the Greeks (Eng. tr., 1925). 

Rosenstock-Huessy, Eugen, Soziologie, Vol. 11, "Die VoUzahl 
der Zeiten'' (Stuttgart, 1958). 

, Zurilck in das Wagnis der Sprache (Berlin, 1957). 

Rousseaux, A., Le monde classique (Paris, 1951). Contains a 
study of Heraclitus. 

Schmidt, P., "Geist und Lehre," Eranos-Jahrhuch, 13 (1945), 
pp. 133-185. 

Sciacca, M. F., Studi sulla filosofia antica (Naples, 1935). 

Scoon, Robert, Greek Philosophy before Plato (Princeton, 


[ 171 ] 


Snell, Bruno, "Die Ausdrucke fur den Begriff des Wissens 
in der vorplatonischen Philosophie," Philosophische Unter- 
suchungen, Heft 29 (Berlin, 1924). 

, The Discovery of Mind; the Greek Origins of Eu- 
ropean Thought (Cambridge, Mass., 1953). 

Stefanini, L., // preimaginismo dei greci: Pitagora, Eraclito, 
Parmenide, Gorgia (Padua, 1953). 

Szabo, A., "Zum Verstandnis der Eleaten," Acta Antiqua 
Academiae Scientiarum Hungaricae, 2 (Budapest, 1953- 
1954), pp. 243-289. Argues that Parmenides' philosophy 
arose by way of reaction to that of Heraclitus. 

Tannery, Paul, Pour Vhistoire de la science hellene (Paris, 


Teichmiiller, Gustav, Neue Studien zur Geschichte der Begriffe 
(Gotha, 1876-1879). 

Vasconcelos, Jose, Pitdgoras: una Teoria del Ritmo (Mexico 
City, 1921). 

Vlastos, Gregory, "Equality and Justice in Early Cosmolo- 
gies," Classical Philology, 42 (1947), pp. 156-178. 

von Fritz, Kurt, "i/ov?, voelv, and their Derivatives in Pre- 
Socratic Philosophy, i : From the Beginning to Parmenides," 
Classical Philology, 40 (1945), pp. 223-242. 

Weerts, Emil, Plato und der Heraklitismus (Leipzig, 1931). 

Wilamowitz-Moellendorf, U. V., "Lesefriichte," Hermes, 62 
(1927), pp. 276-298. 

Wolf, Erik, "Der Ursprung des abendlandischen Rechtsgedan- 
kens bei Anaximander und Heraklit," Symposion, i (1948), 

pp. 35-87. 
Zeller, Edouard, A History of Greek Philosophy from the 
Earliest Period to the Time of Socrates (Eng. tr., London, 
1881). Vol. II, pp. 1-116 of the English edition is devoted 
to Heraclitus. 

[ 172 ] 


abstraction, i3f, 20, 32-34, 42 
activity, see process 
Aeneid, 88 
Aenesidemus, 48, 123 
Aetius, 43, 45, 123, 125, 143 
afterlife, see survival 
air, 6, 40, 47-49, 123 
Alcmaeon, loi, 119, 128 
Alexander (the Aristotelian), 40, 

anachronism, 113 
Anaxagoras, 102 
Anaximander, 4-6, 40, 92f, 113 
Anaximenes, 4, 6, 32, 39f, 46, 113, 

121, 123 
anthropomorphism, 7, 72f, 96f 
appearance, 4-6, 26, 3of, 40 
Aristarchus, 148 
aristocratism, iif, 22, 85f 
Aristophanes, 150 
Aristotle, 16, 20, 33-35, 40-42, 54^, 

59, 62f, 74, 86, 108, ii3f, 116, 118, 

120-122, I24f, 127, 129, 134, I38f, 

142-145, 147, 150, 156 
Arius Didymus, 63, 126, 144 
Asclepius, 91, 121, 128 
athleticism (intellectual), 27 
authenticity, 17, 130-134 
authority, 21, 51-54 
Aztecs, 50 

Bacon, Francis, 7, 20, ii6f 

Bailey, Cyril, 43, I22f 

balance, 3, 51, 107. See also harmony 

banishment, 11, 84 

barbarism, spiritual, 26, 119 

barley drink, 58, 64f 

battle, death in, 86, 127 

beginning, see generation 

Bergson, Henri, 49 

Berkeley, George, 20 

Bible, 132 

biography, 10-12 

birth, 47, 57, 84 

blood, 69 

boundless, cosmic, 5, 93, 113 

bow, 85, 91, 100, 108 

Boyle, Robert, 122 

breath, 6, 63, ii3f 

burial, 11, 137 

Burnet, John, 2if, 56, 85, io6f, ii3f, 

132, 154 
Butcher, S. H., 127 
hyrsa, 88, 128 
Bywater, Ingram, 35, 77, 112, 124; 

App. B, passim 

Carthage, 88 

cause, 4, 20, 39, 41 

ceremony, religious, lof, 69 

Chaldeans, 50 

chance, 35f, 75, 80, 104, i2of 

change, see process 

Cherniss, Harold, 56, 122, I24f 

Chesterton, Gilbert K., 98 

child, 29, 31, 36, 66, i04f, 133 

China, 3, 16 

Christ, 95, 98 

Christianity, 71, 75, 98, I32f 

chronology, ^-6 

circle, 90, 99-101 

city, ^3, 88f, 128 

Clement of Alexandria, 23, 45, 75, 

80, I3if, 134-139, i4if, 145-150, 

154, 156 
clothes, i26f 
coalescence, 11- 16, 26-28, 74, 113, 

118, I44f. See also participation 
Coleridge, Samuel T., 118 
coming-to-be, 30, 47, 120. See also 

common awareness, I9f, 24f, 83, 89, 

community, 9f, 83-89 
concretion, 9f, 31, 40, 42. See also 

condensation, 6, 45 
conflagration, cosmic, 48, 50-56, 8if, 

124, 139 
cosmology, 3, 6, 34-36, 41-43, 47, 61, 

65, 104, 107- 1 10, 119, 124 
Crotona, 8 
cycles, cosmic, 50-56, lOi, I24f, 139, 


death, 11, 20, 47, 74-80, 89, 108, 137, 

dedication, of scrolls, 12, ii5f 

[ 173 ] 


Delphi, 20, ii5f 

delusion, 27, 65, 83, 103, 117 

Democritus, 42, 93, 102 

depth, semantic, 94 

Descartes, Rene, 20 

destiny, 35f, 81, 89, 120 

dharma, 118 

dialectic, 15, 20 

Dido, 88 

Diels, Hermann (including Diels- 
Kranz), 17, 35, I03f, 112-115, 
124, 134, 156; App. B, passim 

Diogenes Laertius, lof, ii3f, 135^, 

143, 145, I48f, I55f 
Dionysus, 69, 88 

distance, spiritual, 8, 11, 85, 115 
downward way, 46, 49, 53, 66, 81, 

86, 125. 
doxography, 17, 63, 130-134, 138 
dream, 20, 27, 78, 137 
dryness, 26, 32, 62, 65, 79, 109, 126, 

144. See also fire 
dualism, 15, 37, 103-105 

earth, 6, 37, 125 

efflux, 46, 123 

Eleaticism, 8-10 

elements, natural, 37, 39» 47f 

Eliade, Mircea, 123 

Empedocles, S^f, 104, 120, 124 

Empson, William, 127 

Ephesus, 10, 84 

Epicharmus, 23!, iigi 

Epicurus, 43, I22f 

epistemology, 13, I5, 31-34, 4h ii7f, 

Erinyes, 102 

eschatology, 61, 66f, 74-81, i26f, 146 
eternity, 30, 35, 54, 75^, loi, 139, 

etymology, treatise on, 153 
Eudoxus, 116 
Euripides, 69, 126 
Eustathius, 115 

evaporation, 45-47, 62, 125, I43f 
evidence, igi, 51-54, 130-134 

Fairbanks, Arthur, 61, 76, 85, 107, 

fire, 9, 25, 27, 30, 37-57, 62, 64, 7% 

8if, 93, 106, i2if, i24f, 126, i4of, 

first principle, 4-8, 14, 30, 33!, 39-42, 

55, 64, 69-71, 84, 87, 92, 96f, 104, 

no, 124 
fools, 58, 83, 85. See also downward 

four elements, 33, 37, 39, 47-49 
Frankel, Hermann, 56, 106, 128, 155 
Freeman, Kathleen, 61, 107, 113, 132 
Freud, Sigmund, 89, 95, 98 

Galenus, 47, 123, i42f, 149 

Galileo, 41 

generation, cosmic, 41, 54, 57 

Gigon, Olof, 56, 87, 127 

gnomic writings, 7 

God, 3, 7f, 57, 69, 71-73, 90, 96, HO, 

gods, 7, 34, 73-76 
Goethe, J. W. von, 14, 44, 118 
gold, 19, 37, 90, 122 
Gomperz, Theodor, 56, 131 
grammar, I3f, 104, 132 
Great Year, 5of. See also cycles 
Greece, 3f, 12 

Greek language, I3f, 43^, 141, I43 
Greek thought, 4, 17, 4 if 
gymnastic style, 91 

Hades, 59, 66, 69, 126 

harmony, 25, 27, 36, 102, 107-110, 

129, 140 
Hastings' Encyclopaedia, 103 
Heidel, W. A., 122 
Helicon, Mount, ii5f 
Heraclitus the grammarian, 152 
Hermadorus, 11, 84 
Hesiod, 3, 46, 84, 91 
Hicks, R. D., 114 
Hippasus, 9, 56, 114 
Hippolytus, 25f, 41, 75, 80, 122, 128, 

134, 136, 139-142, 147, 149-156 
Homer, 29, 46, 52, 83, 124, 129, 140 
Homeric hymns, 3 
horse, wooden, 128 
hybris, 5, 85f 

lamblichus, 9, 137 

identity, 26, 35, loi, no, 120 

[ 174 ] 


idols (Baconian), 7, 9, 13, ii6f 

igneus turbo, 43 

illness, 11 

illusion, see delusion, perception 

image-idea, 14 

immortality, 75f, I26f. See also sur- 

impulse, 49f, 57f, 65 

India, 3, 16, 50 

influence, 8-10, 46, 75, 114 

innuendo, 18, 27, \yj 

intelligence, 15, 25, 41, 68, 71, 102, 
104, 107. See also mind 

interpretation, I2f, I7f, 21-24, 80, 
ii6f, I30-I34» 136, 139, 141 

intoxication, 58, 84 

Ionia, 4-8, 121 

Iran, 3, 16, 50 

irresponsibility, cosmic, 35, ^d, 73, 

Isaiah, 3 
Israel, 3 
Italy, 8 

justice, 5, 13, 29, 68, no, 113, 127 

Kant, Immanuel, 20 

Keats, John, 118. See also negative 

Kirk, Geoffrey S., 24, 44, 51-53, 56, 

79f, 87, 107, 120, I22f, 127, 135, 

knowledge, 19-26, io4f. See also 

perception; truth 

language, I3f, 22-24, 26, 33f, 59-61, 

63, 74-77, 105 
Lao-tze, 3 

Lattimore, Richmond, 61, 107 
law, physical, 49; political, 83, Syi, 


Leucippus, 42, 93, 102 
Liddell and Scott, 87, 138 
life-cycle, 5, 47, 91 
light, 25, 38, 68, 77-81, 144, 146 
lightning-flash, 37, 43!, 123 
Locke, John, 20 
locomotion, 41, 43, 122 
logic, 15, 20, 24, 27, 33f, 41, 63, 105, 

Logos, 19-25, 51, 57, 68, 71, 87, 102, 

107, 117, iigi, 126, 135, 154 
love, 97, 124 
Lucian, 139 
Lucretius, 43!, 123 
lyre, 85, 102, 108, 117, 129 

manifold, 4, 8, 26, 30-35, 4i, 45, 5i, 
113, I24f 

Marcus Aurelius, 109, 129, 137, 146, 

Martin, Everett D., 117 

mathematics, 9 

matter, 4, 39f, 63, 114, 126. See also 
elements; four elements 

Maximus of Tyre, 142, 156 

Mayas, 50 

Mayer, Gottfried, 115 

measure, 42 

mechanical principle, 122 

memory, 70 

metaphor, 25, 71-73, 78, 81, 94-99 

metaphysics, 5, 8, 23, 33f, 69-73, 93, 
103-110, 120, 122, 124, 129 

method, 12, 17, 19-28, ii6f 

Miletus, school of, 4, 6, 8, 39, 45f 

mind, 7, 69, 135 ; cosmic, 41, 44, 49, 
69-71, 73, 102-110 

misanthropy, 11, 84 

moisture, 32, 45, 61-64, 109 

Mondolfo, Rudolfo, 56 

motion, 6, 65, 128. See also loco- 
motion; process 

music, io8f 

mysticism, 124-127 

mythology, 4, 7, 40, I7if 

Nahm, Milton C, 132 

nature, 4-6, 37-57, 93, 116, 120 

Navarro Monzo, Julio, 21 

necessity, 35f, 121 

negative capability, 16, 39, 118 

Nietzsche, Friedrich, 115. See also 

Noetus, 133 
Numenius, 144 

objectivism, 12, 15, 21, 23, 30, 73, 

obscurity, 12, 16, 93, 116, 133 
Oedipus, 127 

[ 175 ] 


ontology, 4-6, 13, 16, 31, 39, 44, 122 
opposites, 5f, 26, 2,3, 120, 140. See 
also paradox; tension 
Origen, 140, 146, 148, 151, 156 
Orphic sayings, 3f, 113 

paradox, 10, 12-16, 26, 33, 66, 78, 

94-98, 100, io8f, 147 
Parmenides, gi, 15, 103!, 115 
participation, 21, 24f, 71, 119. See 

also coalescence 
particulars, 21. See also manifold 
Patrick, G. T. W., 151 
Paul, St., 114 
Pausanias, 116 
perception, 6, 26, 31, 59, 63, 65-67, 

118, 138. See also epistemology 
periodicity, see cycles 
perspective, 5, 27, 46, 62, 82, 86, 121, 

pessimism, iif, 115 
Pfleiderer, Edmund, 108 
Phaedo, 76, I26f, I28f 
Philo, 137 
philosophy, idea of, 23f, 32. See also 

truth; wisdom 
physics, 6, 37-57, 120, 122, 124 
Plato, 3, 9, 16, 20, 27, 52-54, 76, 93, 

123, 127, I37f, 150, I55f 
Plotinus, 139, 149, 156 
plurisignation, 78, 127 
Plutarch, 45, 47, 56, 120, I23f, 137, 

140, i42f, i45f, i48f, i52f, I55f 
pneuma, 118 

Poetics (Aristotle), 127-129 
Polybius, 136, 156 
polytheism, 7, 74f 
Porphyrins, 144, 15 if, 156 
potentiality, 5, 113, 126 
prayer, 69f 
process, 3-6, 13-15, 27, 29-36, 63, 75, 

93, 105, 120, 125, I38f, 142; et 

Proclus, 137, 156 
prophetic role, 22, 69 
Pseudo-Aristotle, 41, 112, 116, 125 
pun, 27, yy, 127, 147 
purity, 122, 155, 
pyramids, 42, 122 
Pythagoras, school of, 8f, 27, 84, 

93, 107-109, ii4f, 119 

qualities, 5f, 13-15, 31-34, 42, 65, 

quantity, 6, 122, 141 

rarefaction, 6. See also upward way 
Raven, J. E., 24, 120 
reason, see Logos; mind; intelli- 
religion, 10, 16, 68-82 
river, 29f, 37, 90, 142, 152 
Rodin, Auguste, 39 
Rouse, W. H. D., 116 

Schleiermacher, Friedrich, 14, 17, 

135, 160 
Schopenhauer, Arthur, 11, 19, 115 
science, 4-6, 32, 37-56, 119 
sculpture, 39 
self, see soul 
self-assertion, see hybris 
self-examination, 19, 27 
semantics, 13-17, 31-34, 77^ 122, 127. 

See also paradox 
Seneca, 43, 149, 152 
Sextus Empiricus, I34f, I36f, 146, 

Shakespeare, William, 118 
Sibyl, 69 

Simplicius, 40, 42, 56, 74, 140 
sleep, 27, 69f, 78f, 129, 137 
smell, 59, 66, 126, I45f 
Smith, J. A., 126 
smoke, 59, 66 
Sophocles, 15 
soul, 6, 59-69, 7of, 75-77, 79^, 109, 

119, 126, 129 
Spengler, Oswald, 37f, 121 
Spinoza, Baruch, 20 
Stobaeus, 119, 124, I35f, I44f, I47f, 

150, 156 
Stocks, J. L., 56, 124 
Stoicism, 50, 54, 109, 123 
Strabo, 113, 143, 146, 150, 156 
strife, 35, 53, 120, 133 
subject-object, see epistemology 
Suidas, 114 

survival, 74-80, 118, 126, i46f 
symbol, 14, 30, 38f, 66, 122 
Symmachus, 149 
synecdoche, 44 

[ 176 ] 


tao, no, ii8 

Tao Teh Ching, i6, no 

Teichmiiller, Gustav, 39 

tension, 10, 27, 85, 104, 108, 140 

Tertullian, 116, 143, 151 

Thales, 4, 39f, 45, 74, 92f, 126 

Theaetetus, 2 if 

theater, idols of, 13, ii6f, 122 

Themistius, 137 

theology, 69-74 

vanishing, 29f, 120, 142 

Vergil, 88, i27f 

Vita Homeri, 149 

von Hartmann, Eduard, 49 

wall, city, 88, 128 

Walzer, Richard, 132, 134; App. B, 

war, see strife 

Theophrastus, 39, 65, 126, 143, 145, way up and down, 6, 45, 80, 108^ 

time, 5f, 30, 36, 5of, 139 
tragedy, 86, 127, 129 
transcendence, 7, 23f, 73, 102-110, 

119, 129 
triadicity, 33f, 120 
Trinity, doctrine of, 133 
Troy, 88, 128 

truth, 21, 23-25, 93, no, 154 
Tzetzes, 139, 150 

unguents, 106, 155 

universal, 20, 22, 93 

Upanishads, 3, 16 

upward way, 6, 12, 49, 53, 81, 86, 

102, 109, 125, 128 
usurpation, 127 

139. See also downward way; 

upward way 
wisdom, 3, II, 15, 20, 23f, 38, 7of, 

Word, see Logos 
word-magic, 24 
world cycles, see cycles 

Xenocrates, 116 

Xenophanes, 7f, 41, 72, 114, 126 

Zarathustra, 3 

Zeller, Edouard, 10, 56, 125 

Zend-Avesta, 116 

Zeus, 69, 102, io5f, no, 120 

Zoroaster, see Zarathustra 

[ 177 ] 


Page numbers are indicated by roman type and Fragment 
numbers by italic. 

ayaXfjia, 75 

dyaOos, QI, QQ, I06 

dyup, gs 

aSiKos, 106 

del, 138 ; 29, 43 

o-VP, 123; 34 

dddvaros, 66 

"AidTjs, 'Atdris, 59, 77 

atfLa, 78 

aipiKTTjs, 116 

aiaOrjTos, 1 38 

atriov, 121, 122, I41 ; 30 

ai(av, 139; 24 

dKOTf, II 

dXrjdeia, I28 

dX'qdris, 10 

dfxaeiy], 53 

dfiaprdpio, I29 

dfidpTTjua, 127 

dfiapTia, 86, 108, 127, 129 

d/xotp-fi, 141 

dvdyKT], 36, 120 

dpaOvixiaaris, 45, 62, 143-I44; 43 

dvaOvixidoj, 44 

dpairavofiai, 23, 97 

dpairavais, pp 

apa^, 18 

dpeXiriffTOS, I3I, I37, I38 ; IQ 

dpcfx/xa, 142 

^^VP, 3, 48, 95, 105 

dpdpojireios, 61, 81 

apOpuTTOs, 25, 29, 62, 65, 67, 69, 86, 

loi, 104, 106 
dPTafioi^Ti, 141 ; 28 
dpri^ovp, 98 
dvu), 128; 108 
d^vperos, 55 
doidos, 91 
dirapra, I25 ; 28, 29, 85, 94 

dTTCipOP, TO, 113 

diriaTlr], 63 
diTTU), 68, 78; 65 
dpeTT}, 10 

dpiffTos, 86; 46, 84, 85 

apKTOS, 39 

ap/iopia, 107, 129; 98, 116, 117 
dpfio^oi, 153 

dpxv, 40, loi, 113, 121; 109 

'ApxiXoxos, 93 
daut/xarwraTos, 43 
druxTj/xa, 127 
avyrj, 144 
avos, 46 
d(f>apris, 116 

pdKxos, 76 
pdp^apos, 13 
/3acrt\eus, 25 
^affLXrjir], 24 

/Si'os, 100, 153; 66, 115 
/366s, 100, 153 
/3Xa^, 54 
^ovXri, 83 

yepeffis, 120 

717, 3^, 33, 34, 49 

yiypofiai, 124, 125 ; I, 26, 49, 52, 98, 

ypd<f>eios, III 
ypwixT], 61, 120 

8aiixu}v, 69, 105 

AeX0oi, 18 

SrjiMos, 82, 91 

dia<pepoi}, 129 

diKaios, 106 

diKT}, 113; 26, 71, 100, 122 

dLoUifja-ts, 122 

Atopvaos, 77 

5o/cew, 57, 67, 87 

SoKifxos, 87 

56|a, 128 

dovXos, 25 

dvpop, 73 

iyeipo}, 1 6 
iyprjyopeo}, 15, 65 
eiKT), 5 
ei/xapfiivT], 120 

[ 178 ] 


clvai (used existentially), 115, 124; 
I, 2, 2g 

eiprivT), 121 

eKiroia, I23 

€Kirvpu}(ns, 50, 52, 53, 54, 124 

iKwvpovfiai, 124 

eXiro/xac, 67 

eXTTTjTOS, 131 

iXiri^io, 132, 137, 138; 19 

€IX<j)pO)Vy 126 

y, 128 ; 8s, 84, 112, 114 

evavrios, 14O 

eVe/ca, 143 

'ivvXos, 126 

e'lai/^is, 143 

e^evpiaKio, 1 38; Jp, J^^ 

iirivoia, 122 

e/ots, 140; 26, 9<? 
'Epfiodopos, 95 
evdu), 16, 65 
evdvs (adj.), HI 
evxofiai, 75 
'E<f>eaLos, 95, 96 

Zeus, 120; S9, 119 

ijdovri, 121 

riOos, 61, 69 

vXlos, 36, 37, 3S, 122 

i}p.epa, 36, 94, 121 

ripuis, 75 

'HcrioSos, 94 

edXacraa, 32, 33, lOI 

Odvaros, 16, 34, 47, 49, 66, 118. See 

also 67 
delot, 128; 61, 63, 81 

Beds, 25, 29, 126 ; 74, 75> 79, 86, 104, 
106, 121 

dvTJTOS, 66 
eVflOS, SI 

dvufia, 155; 121 

larpos, 107 
1810s, 15 
iepos, 56 
Ix^vs, lOI 

Kadaipu), 78 
Kadevdu}, 14; 124 

KOLfxaros, 89, 99 

Kairvos, 58 

Kap4>aX€0S, 22 

KaTaXafjL^dub}, 71, 7^ 

Karcj, 108 

Kepavvos, 123 ; 35 

KLveofiat, 43, 50 

KLvrjffis, 122, 126 

/cXeos, 85 

Kotvos, 135 ; 2, 15 

Koirpos, 60 

Kopos, 30, 99, 121 

Kofffios, 122, 124, 141 ; 13, 29, 40, 124 

Kpiv<a, 72 


KV^epvato, 120 
KVKecjv, 50 

KiKXos, 128 ; 109 

KVOJP, 90 

Xiyo), 21, 119, 120, 138; 18 

Xi/x6s, 99, 121 

XoyiKos, 62 

Xoyi<rp,6Sf 120 ^' 

\6yos, 21, 22, 23, 63, 92, 118, 119, 

120, 126; I, 2, 7, 33, 42, 45, 54, 

64, 118 
X{>pa{-rj), 117 

fidyos, 7<5 
fiddrjais, II 
fxalvofiai, 78, 79 
fiavrelov, 18 
fidpTVp, 12, 13 
p.€(ra, 113 
fiera^dXXci}, 23 
fiirpov, 141 ; 29, 122 
fjLLffdos, 107 
fiotpa, 70 
fiopos, 70, 97 
fiveoj, 76 
fivarriptov, 7<5 

vcKpos, 68 
piKVS, 60 
vriTTtos, 105 

VOCTOS, 126 

pofios, 87 ; 81, 82, 83 
voos, 120 ; 6, 81, 91 
voaos, 5^ 
vovaoSy 99 

[ 179 ] 


ivv vw, 120; 81 

ivvos, 100, 120, 134, 144; 2, 26, 80, 

656s, 108, III 

0L7](TIS, 56 

oIkc^co, 35 

olvos, 53 

oh, (ara, 12, JJ 

oKos, 122 

"Ofivpos, 92, 95 

ofioXoyeu, 154 ; I18 

'6v, 48, 123. See also ehai 

'6pofia, 100, 115, iig 

ovos, 102 

ovffia, 121 

o^daXfios, 12, IS 

6xfyis, II, 65 

irddos, 126 

Trais, 24, 48, 97, 105 

irakiv, 124, 126; 31 

irakivTOvos, 154 

iroKLvrpoTTOs, II5, 153; H? 

irav, irdvra, II3, 124, I25 ; I, 7, 20, 
25, 26, 28, 35, 41, 58, 72, 98, 112, 
118, 120, 123. See also diravra 

Trdvra dia iravra IO4; 120 

irdvTes, 80, 81 

ireKoj, 115 

irepas, 10 1 ; lOg 

Treyote'xw, I18, I28 ; 62 

irepL<f>€p€ta, 109 

iridriKOS, 104 

iTKTTevb}, 132 

■K\7iyf\, 41 

irXovTos, 96 
TTvevfia, 114 
TToXe/AOS, 120, 140; 26 

TToXtS, 81 

TToWdy rd, 63 
iroWol, ol, 2, 57, 85, 91 
TToXvfiaOlr], 6 
TTOfiTrri, 77 
TTOPTjpevofiai, 96 
irora/xos, 21, IIO, 121 
Trpoeiprjfievov, I18 

TTvp, 122, 124, 125, 141 ; 28, 29, 32, 
34, 72 

pio), 138 ; 20, 43 

ptP€S, 58 

adpfia, 40 
(T^evvvpLi, 88 
ff^eais, 143 

anjixaivta, 137 ; 18 

St/SuXXa, 79 

<7Ko\i6s, 99 ; III 

<tkot€lp6s, 116 

(ro<pia, 104 

ao^os, 23; 7, 46, 118, 119, 120 

CFTOLX^tOV, 113 

av/nfievo), 126 
<rvfX(f>ep(a, 112 
(Tvvq,5ov, 112 
avvd\f/i€S, 112 
avvepySs, 124 
crvj/iffTTj/iii, 124 
avpfia, 102 
trwfw, 126 
crwfia, 122, 124 
crcafiartKos, 121 
aoj^poveu}, 136; 9, 10 

relxos, 82 
Tekevrri, II3 
TeKo^, lOI 
ri(JLS, 113 
Toiov, 115, 117 
Tvpavvos, 127 

v^pis, 88. See also hybris in General 

vyLetT], 99 

vypos, 22, 44, 47, 48 
vdcap, 34, 49, loi 

vXlKOS, 121 
V/JLVOS, 77 

0dos, 65 

(pOelp, 92 

<j>dopd, 120 

<(>tX6(ro<pos, 23, 135 ; 3 

(ppevr}p7)s, 62 

(fypoveu), 136; 80 

4>p6pri<7ts, 135 ; 2 

^povLfios, 122 

(f>priv, 91 

<pvXa^, 68 

(pvaioXoyia, II3 

0UO-IS, 113, 134; I, 10, 17 

<t>vb3, 113 

[ 180 ] 


Xei/AWJ/, 121 i^vxr), I3> 4^-49, 5^, 59 

X/oewj', 26 \{/vxp6s, 22 

XPVfiara, 28 

Xpr]<^fJ'0(rvv7], JO ojpa, 12^ 

Xpvffos, 4, 28, 102 wra, see oh 

[ 181 ] 

Date D 



Aljp ? '^ IQ^fl 

Nuv 2 4 m 





Heraclitus. main 

3 lEbE D357T flbTS 


Date Due 





» r" 

■ ^ 







^^ ■■ "^ 







^ \ 


m„ !.